Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?

FreeMWare: Like VMWare but Open Source 303

CentrX writes "I was surprised that no one has contributed a story about FreeMWare since they started. FreeMWare is "an extensible open source PC virtualization software program which will allow PC and workstation users to run multiple operating systems concurently on the same machine." Like VMWare, only free and open-source. They now have a CVS repository and the latest source can be downloaded. I think this project is needed and needs some support from the community. You can also join the mailing list." FreeMWare was mentioned briefly here in April. Looks like it's come a long way since then.
This discussion has been archived. No new comments can be posted.

FreeMWare: Like VMWare but Open Source

Comments Filter:
  • RC3 runs just fine on VMWare 1.1
  • Sorry, I don't see it. It seems to me that paying for support for free software is exactly the wrong economic incentive to create software.

    The company in question profits almost completely by releasing a buggy, badly documented product.
    Nonsense - because:

    A) No-one will use it

    B) Depends on the support contract. If it's per incident then maybe, but most I know of are of the yearly contract form. Then the less bugs in your product, the less support staff you need, the more money you make... Ipso facto.
  • Not to complain or anything, but how does that post qualify as "Redundant"? I just looked through every other post in the forum and failed to find *anything* which made the same point that I did (admittedly, Yes, there wasn't an incredible amount of rhetoric or analysis, but there isn't a "Shallow" button in the moderator widgets.

    To repeat myself (now I guess I am Redundant ;-) ): 90% of the people who are screaming bloody murder because Freemware is "ripping off" the VMWare folks are apparently running Linux. Linux is a free clone of the Unix[tm] operating system which at the time was being sold by various commercial companies. Linux has no particularly innovative features; in fact, it's still catching up with Solaris in many areas and its feature list is essentially borrowed from other operating systems. Its distinguishing characteristic is that (drum roll) it's free! In fact, it was originally written because Linus Torvalds wanted to use Unix on his PC but didn't want to pay for the commercial Unix systems. Does this sound familiar yet?
    In other words: the same people who are up in arms about Freemware ought logically to be up in arms about the very operating system they are running it on! Anything less is inconsistent at best and hypocritical at worst.

    Hope this makes my point clearer,
  • Umm, how is VMWare a competitor?

    It's a competitor in the sense that someone with an X86 system is unlikely to need or want both VMWare and Bochs.

    So he's giving his time to a cool project. It's his motives I question. He went to VMWare's news server and tried to recruit people for his VMWare clone right there. That's rather obnoxious.

    His major criticism of VMWare was that it was not open source, yet Bochs is not open source. If he can blast them for not being open source, I don't see any reason I shouldn't blast him. If I wanted to make the correspondence perfect, I suppose I'd have to go blast him on a Bochs mailing list.

  • Performance ceases to be an issue above 192M RAM

    My current machine is a PII-450 with 192M RAM, of which I ususlly keep 112 for the host and give 80M to the guest. They both seem happy with this setting (96/96 was a bit unfair to the host), and the speed is good. The only time when it _really_ slows to a crawl is when I access the external parallel zip drive directly, via the virtual port, bypassing Linux. It can take 30 minutes to copy 95M of files, and the host OS becomes unusable all along. Easy to figure I don't do this too often :)
  • Cool?

    How about those poor bastards at VMWare? They will surely loose a lot from this.

    Open Source has its dark side too - since it rips programmers from their jobs. If this Open VMware becomes successful, there is a chance that VMWare will not pull in big enough revenue to support all of its programmers!

    And we all know that VMWare is not our enemy, after all it DOES innovate and does produce some amazing software for Linux. Why would we want to hurt a company like that??

    Not Cool.
  • If they included some pluggable module architechture, here a three examples that come to mind:

    1. Various emulation modules that would allow you to run different architechtures on your x86, or run FreeVMWare on architechtures different than x86.
    2. A trace module that would allow you to see direct interaction with every component in the system; useful for hardware and software developers.
    3. A passthru module that would allow your VM direct acccess to a real piece of hardware. This would be handy when writing drivers for things; hose your host O/S? Reboot your VM!

    Of course, you would be able to stack the trace module and the passthru module together for a bitchin' development environment!


  • However, for whatever reason, it needs a lot more RAM. It has to physically allocate however much RAM you tell it to use for the emulated OS, in my case 32MB for Win '98[...]

    Well... Win98 is a monster in its own right--I'm surprised that you can get it to run at all in 32MB :(. The reason VMWare wants fixed blocks of RAM is that tricking an OS into thinking it has complete control over memory that is actually managed by another OS is a complicated process, and while they probably could do it in a more dynamic fashion (only allocate physical storage for memory the guest OS actually uses), that would probably significantly increase the complexity of the code, which of course usually means increased bugginess and performance penalties.

    DOSEmu, by contrast, never uses as much RAM as I tell it it can use unless it absolutely has to. Usually I give it 8MB, but when I wanted to run Callus, I gave it 20MB. Worked great, except for lacking sound. Wine generally uses 4MB[...]

    DOSemu has a much easier job than VMWare because hardware support for x86 "real" mode virtualization is built directly into all modern Intel class processors. DOS programs also rarely use more than ~1MB of memory in any case.

    WINE, of course, "Is Not an Emulator" ;). It's basically just a library that relays Win(16|32) API calls to their corresponding xlib calls. The windows apps themselves are executed directly with the help of the loader program, so it's actually a bit surprising that it uses even that much memory.

  • That's all i would really care about at the moment. If the open source version is as good the the real deal, then I'd recommend that any shop that could use a spare machine(s), that wanted to allow their developers to build and test client/server applications on virtual machines connected by a virtual network on a notebook computer (or regular desktop), or had more than 1 type of desktop to support (everybody?)!

    You get the idea, VM's (whoever cooks 'em), facilitate support and experimentation. This is good for people and companies looking for alternatives on the computing landscape. Looking to begin using open source software on their machines! Looking to preserve their investments and use the best software for their needs.

    Credit to both teams. This is terrific stuff.
  • So of course it won't work properly on every platform yet. That doesn't mean that the project is a failure. Just the fact that it *compiles* on BeOS at this stage could be seen as a success.

  • You said "commercial" software is greeted with hostility. Don't you really mean proprietary software?

    So how can free software be commercial ? And don't tell me about Caldera/SuSE who rely on proprietary software, or Redhat, who aren't making money.Is there any evidence that free software is a money maker ? It seems to me that a lot of people care more about the beer than the speech.


  • More specifically, it raises serious questions about whether proprietary linux companies can survive.

    I don't see the need for "more specifically". I made and meant it in full generality.

    Just what the hell are you talking about?

    I am talking about the fact that a lot of people come to linux for the beer and not the speech.

  • by weave ( 48069 ) on Sunday December 05, 1999 @08:13AM (#1478775) Journal
    While I don't have any idea whether it's technically a trademark violation, I wish the developers had chosen a name that wasn't simply a variation on VMWare.

    A little history, to be fair. The idea of a "virtual machine" and the name VM came from IBM a looong time ago, 70s I believe, at least well before 1980.

    IBM sold VM that could load and run other of its operating systems on 370-based mainframes, like MVS or even other copies and versions of VM. This was very helpful for system administrators. Back then, you just couldn't go buy a test box for trying out newer versions or for testing risky patches, etc. So you just ran another guest OS under VM!

    So, if anyone has a beef about using the VM acronym in a product name describing this feature, it should be IBM.

  • Wow

    I've been thinking about buying a license for VMWare (but since I don't have a lot of money, I've delayed it over and over) and now there suddenly is an open source project. That's just the best I could dream of. I'm looking forward to the first final release.
  • May be you read more into my message than is there, or may be I didn't make myself clear, which is more likely. I know that you can run VMware on a SCSI only system. But, I also know that you can't run it on raw partitions. You have to create a file on your Linux partition that acts like the Windows hard drive. Yes, I would like all the raw power of U2W to be available to guest OSes. Making a dual boot system defeats the purpose of having VMware, so that point is moot.

    As I currently understand it, I can't use a partition on my SCSI drive as a raw partition under Windows.
  • Well don't expect a usable release any time soon. This looks *very, very* alpha. And I guess writing a complete VM is a hell of a lot of work (I'm just checking out the paper on "virtualization"). But it *is* cool, and choice is always good.
  • Yes!
    I like this way of putting it.
    It hilights that it takes time to develop programs. This is a lot like the idea of licensing things with clauses making the software public domain after a certain period of time.
    I wish legislation would be passed limiting ALL software liscences as to time. After it's not making money for the creators anymore, the public should have it.

  • by G-funk ( 22712 ) <> on Sunday December 05, 1999 @03:02AM (#1478785) Homepage Journal
    I've been on the mailing list for freeMWare since it started, and although I'm quiet, I pay attention. It really is coming along, and is (unlike other group projects I've followed) actually being led somewhere by the remarkable mr Kevin Lawton (sp?). I have to say, congrats to all the contributors, as this is a big step in the right direction.

    -Josh G
  • So the natural question is, where is the time limit? How do we make sure companies (cf. Microsoft) do not make excessive amounts of money, more than is a valid reward for a small amount of innovation?

    When was the last time a proposed open source project was cancelled because the commercial version's vendor was too small?

    It seems that some people feel *any* amount of money is excessive.

  • No, but I for one would have confidence that if there was a trojan in there, somebody would notice it and point it out, and it would probably show up here on /., for all the world to see.
  • Ofcourse. I think this could really work out, because I think there's a lot of need for a good system like this. Ofcourse, VMWare has THE market share now, but that could change. Look at microsoft and linux .... :-))
  • I agree 100%. But this also raises some serious questions about whether linux companies can survive. Because to date, it seems that any attempt on part of a linux company draws hostility from large factions within the user base.

    In particular, commercial software is invariably greated with hostility, and the birth of a project whose sole aim is to do a "cheap imitation" of the innovative commercial product, which inevitably will have the same kind of effect as IE had on Netscape. Linux hardware shops are often turned down by linux users in favour of windows-only shops. Linux users don't seem to vote with their wallets.

    However, I am still hoping that, like you and I, there will be others who will pay their fair share. I am saddened that there is a faction of linux users that remind me of the warez scene.

  • Cool?

    How about those poor bastards at VMWare? They will surely loose a lot from this.

    Only if they can't produce a superior product.

    Open Source has its dark side too - since it rips programmers from their jobs. If this Open VMware becomes successful, there is a chance that VMWare will not pull in big enough revenue to support all of its programmers!

    And we all know that VMWare is not our enemy, after all it DOES innovate and does produce some amazing software for Linux. Why would we want to hurt a company like that??

    If they fail to produce anything worth throwing any money at, why should we continue to throw money at them? If they succeed in continuing to produce something worth paying for, we'll continue to pay for it.

    Open Source does not rip programmers from their jobs unless they have worthless jobs. As long as they're producing something that there's a good reason to throw money at, people will continue to throw money at them. I've already thrown them my $99, and the company I work for will probably throw them $299 early next year, once I convince them of the advantages their product provides. And as long as they continue to provide software that enhances our computer use, we'll continue to provide financial support for them to do so. They're only going to stop getting money if they fail to provide anything beneficial to us, which is as it should be.

    I look forward to FreeMWare, because I suspect it'll spur VMware to produce an even better product in order to stay ahead. They got a good head start but they'll have to work if they want to keep it, rather than just rest on their past success. So, as a paying VMware customer, the FreeMWare project benefits me, even if I never run FreeMWare's product. I'm glad they're out there, and I wish them luck!


  • Hopefully, this idea can be extended far further than it is today. Imagine the ability to use a single generic machine (x86 or otherwise) that can emulate a variety of systems, including NT, Linux, BeOS, Sony Playstation, and your intelligent toaster. It's certainly possible, but it's a lot of work. Perhaps this is the first step. It's interesting to see where this may go. I'm interested in seeing what kind of performance they get. I'm also interested in how they manage to split the code between running natively and running emulated.
  • by ShadowDragon ( 40886 ) on Sunday December 05, 1999 @03:05AM (#1478792) Homepage

    I for one am glad this is in development. I used VMWare back in it's beta days and was quite impressed with the idea. I wasn't that impressed when I started getting spammed about the release version and 'send us x ammount of dollars so you can still use this.'

    If I was going to pay them the ammount they wanted, I would expect that all of my hardware would work with VMWare, but it wouldn't recognize my windoze partition and made me re-install, wouldn't let me have the 6 IDE devices I have in my pc (4 HDD, 1 CD-ROM, 1 CD-RW)

    Now with FreeMWare, it's free, I expect stuff like this, and spending hours configuring it to be useful. I wonder how this will affect VMWare's pricing scheme?

  • $100 for personal use. $300 if you actually do any work on your computer.

    I'm torn up about this because it is more simple and more cost effective for me to go out and buy a small second system and outfit it with a monitor/keyboard switchbox.

    The real power is in creating multiple machines on one for experimenting with system interactions over networks. But you still pay through the nose for memory.. and there are a lot of things you still can't do.

    And are games even on their "to-do" list?

  • There are always a few non-free pieces of software on someone's box, and one that I've notices a lot of people using in VmWare. Along with Mozilla replacing Netscape, pretty much the only non-free sw people will be using soon will be Q3A [well, beside the OS they will be virtualizing, of course]

    Seriously, this is great news for everyone, and I wish the developers good luck with their efforts. I'm looking forward to a release: in fact, if I had more than 64m of memory, I'd go try out the unstable version for kicks (who needs uptime, right?)
  • 1) You completely misunderstood what I said. There are tons of tradeoffs in CPU design, and lots of ISAs is fine with me. The point was, if I have the source, I can recompile for whatever I happen to use. A transmeta style CPU will of course never run a given ISA as fast as a CPU tweaked out to run one that specific one.

    2) How many mainframe apps would be useful to run on your desktop PC?

    All of the programs I use regularly are open source. And as more and more open source apps come out, more people will be able to say that.
  • You can download and even redistribute the source, but you can't modify it. The license [] does imply that you can modify this source for internal use only, but that isn't explicitly stated. You can apparently send modifications to the author, who might incorporate them into future releases.

    Anyway, Bochs is not free software, in either the beer or speech sense. The "you can look at our code and send us suggested changes" reminds me more of the SCSL [].

  • I know that there are functions which I've seen which served no apparent purpose.

    True, sometimes other peoples' code looks like it was written in some sort of, well, code. :-) If I'm just browsing, I won't bother to try to unravel it. But if I'm working on the same program, I have to figure out what it's for. Well, actually, I probably don't have to most of the time, but I want to.

    The general case when dealing with other peoples' code is, if you don't understand something, leave it alone, which I'm pretty sure applies to most of the kernel for most people.

    Umm, no. Don't leave it alone. In fact, messing with it can be one of the quickest and easiest ways to figure out what it does. Add a line that prints something when it's called. Comment out the entire body and see what the program doesn't do right without it! Put in some debugging breakpoints, trace it, throw in some ASSERTions, making guesses as to what conditions obtain while the code is running. In short, just play with it. Frequently that's the easiest and most effective was to discover what some code is good for.


  • >The point was, if I have the source, I can recompile for whatever I happen to use.

    Its not always that easy. There are great mountains of bad and machine specific code out there. Sure if everything were open source and ANSI/ISO it would be just fine, but a great many systems are just braindead in the way of portability.

    >A transmeta style CPU will of course never run a given ISA as fast as a CPU tweaked out to run one that specific one.

    True. But a ISA-morphing CPU will run many orders of magnitude faster than software CPU emulation. Would you rather have one machine that can run many ISAs or a seperate machine for each ISA out there?
  • I'm curious, there's Open Source software movement, why isn't there an Open Source hardware movement. It would be harder, but possible I think.

    I'm not sure what the point would be. With open source software, I get the blueprint for my software, which I can modify and recompile. With "open source hardware", I would receive the blueprint for my hardware, which I could could modify, but then what? I'm not about to set up a factory and sink a few million dollars into producing my tweaked system! Back in "The Goode Olde Days(TM)", I expected and did in fact generally receive blueprints of my computers, and did modify them with off-the-shelf parts, but the days of the mighty 16K computer are long gone. I no longer have schematics for my current PC. I'm not sure what I'd do with them if I did...


  • Note that neither of these points apply to software. Software can be enjoyed by infinitely many people simultaneously. Software runs on MY computer in MY house. No one elses property is involved. And don't go off on that 'information is property' crap.

    It's not the information, dickhead, it's the labour put into the software. Do you work for free? Who pays for your food?

    What would make sense, possibly, is id says "we have this idea to make a good game. It will be great. We will make it when we recieve 10 million dollars."

    And in the meantime, the programmers work at McDonald's? Or live off Social Security and soup kitchens?

    Well, if you don't pay, you may not get anything. Gamers pay now, even though they could download it warezly. So they would certainly pay under such a system.

    You should join the Olympics team: With logical leaps like that, the long jump world record is well within your reach. The two are not comparable.

    If you cannot provide good arguments, and experimental evidence that proprietary distribution is the only software model by which developers can be adequately compensated

    It's not, but it's one of many models. So you prefer some other model - but why do you feel the need to attack those that treat their work as more than a mere hobby? Does your model not stand up to competition? It starts to sound like a religion...

    you are a bastard in my book if you make use of this evil system.

    *sigh* There you go off on a religious tanget again.

    It comes down to whether you consider manufacturing software a labour or a hobby. If you work, would you do so for free? It could be argued your labour is not a physical entity - why should you get money for it?

    If some pay for that labour and others don't, it makes those that don't pay (the warez kiddies) leeches, in the sense that they receive a benefit they haven't contributed to the availability of.

  • Ive been thinking about the same thing for quite a while, but I just dont have the resources to start a movement like that.

    What I think would be interesting is for someone to make a CPU out of a PAL device and open source the device-description file. People would then send in "patches" to the chips hardware architecture itself.
  • The question begs asking though: if they didn't want the perception of being VMWare wannabes, why make the name of their project imply that they are?
  • You're pretty lucky to even have a computer. I have friends that can't afford a computer at all, much less get free stuff second-hand. You make it sound like you're living day by day, never knowing where the next meal is going to come from.

    Come on. You're sitting around on the internet, wasting time on a web site! You've obviously got a house, a computer, and phone service. Do you think that everyone in the country has all that? Do you think that *maybe* you take what you have for granted?

    I know what it's like to eat Spaghettios every day for lunch and dinner for a month, but, really, do you think that you and me have ever had it as bad as someone who lives on the street? I hardly think so.

    What the hell does Richard Stallman's ideals do for the people who are shivering and freezing to death in their tiny apartments, because they can't afford to pay their heating bill?

    Let's get a little perspective here, folks, before we start going off on the "I'm soooo poor... oh woe is me... I can't afford to buy VMWare..."

    Yes, having a free counterpart is a *good* thing, and I commend the programmers if they are able to finish such a massive project, but let's view it like it is: an act of charity, not someone saving the world!

    I like free stuff as much as the next person, but when I can't afford something, my first thought isn't, "My rights are being trampled upon!" Rather, it's more like, "How can I go about getting this product or functionality while staying under budget?" Sometimes you're lucky and there's a free version. Sometimes you have suck up and deal. Or do it yourself. That's life.

    I'm lucky enough to be upper middle class now and not have to worry much about necessities, but I also know what a necessity actually is. VMWare is hardly a necessity. It's a very cool program that makes life easier. Beyond that... well, you'll live without it.

    I appreciate your situation and understand that you depend on free software, but talking about "survival", like you *need* some computer program to live, is ridiculous.
  • No, but I for one would have confidence that if there was a trojan in there, somebody would notice it and point it out, and it would probably show up here on /., for all the world to see.

    Bingo! It's the "many eyeballs" effect.


  • I first noticed this project whilst my continuing evaluation of VMWare was taking place back in April. The site was pretty empty back then, but they had a great idea - produce an open sourced, free (as in FSF definition) vitual machine running under Linux. Basically one can have another "machine" running on top of whatever they have already. The system works because the "guest" operating systems are originally written for the processor upon which the virtual machine is running. The virtual machine runs as a conduite or "bridge" between the physical processor, the OS and the guest environment, passing processor instructions on to the main processor. Several problems can occur with this.

    1). You can only run an operating system which was originally developed for your particular processor - in the case of the x86 family of chips, this just happens to include a wide range of OS's (Linux, FreeBSD, windows, dos, Minix, etc.) so we are quite fortunate that we can have the "facilities" provided by certain alternative operating systems. What we can't do is suddenly run MacOS on our PII. Transmeta may or may be not working on software to allow chips in the embedded market out of this situation when it comes to embedded OS's

    2). You have to create an actual "mock" machine for the OS to run inside of. In the case of VMWare, they use a Phoenix BIOS inside the software to provide a basic functionality for the system. The BIOS provided is often very basic in its construction and provides only limited features.

    3. You have to provide third party vendors access to your underlying hardware through device drivers. You must be able to run a device driver from your graphics, soundcard, etc. vendor on the VM without any glitches. All the obscure "trade secret" NDA oriented code must just work, including any obscure ways that it accesses the cards inside the system, any broken implementations that it takes advantage of, etc. The PCI standard for one is known to be poorly compliant in implementation with its own standards - all such wierdness must be replicated entirely.

    VMWare have produced some reasonable software. True it's commercial in nature, but it does provide some degree of OS independence. Ufortunately, VMWare will not yet support some of the really cool stuff, such as DVD access for playing DVDs. It is also quite expensive (more expensive even than windows).

    Freemware is going to be really cool. Unfortunately, and now getting back to the original point of my post - Freemware do not have access to some of the key elements required to build a virtual machine. They need to provide the physical layer that conceptually sits atop the uderlying kernel/hardware and provides kernel/device access. They need to provide this access through an emulated BIOS. VMWare have the Phoenix BIOS whcih they have bought in for use in VMWare - Freemware have to start from scratch. Furthermore, companies such as Microsoft (and their "supporters") are bound to be unhappy with the ease of which an OS could be monitored/reverse-engineered with such technology. As such, you can bet they make it as hard as possible for such machine emulation to be successful. Some companies aren't going to like the idea that people are running their stuff on emulated machines - their support droids won't know where to start when diagnosing some more complex problems.

    Freemware have to emulate pretty much everything inside a standard PC. Does anyone actually have a "standard" PC? NO. More precisely Freemware need to emulate all of the little quirks inside modern computers - the "slightly broken" standards, the increasingly wide variety or hardware and the drivers for the emulated hardware - linked to the physical hardware via lower level kernel calls and raw device access.

    All the while, they have to fight the age old battle against the large corporations who don't want people to know how their stuff actually works. I don't actually know anyone who can describe in detail exactly how the entirity of a modern PC works. Freemware have a hell of a task ahead of them.

    Remember, companies like VMWare have bought in the core technology from others and wired it together using their own code. Freemware have zero co-operation (probably the opposite) from large corporations and they have to do everything from scratch.

    An interesting question is "does windows 2000 run properly on VMware?" Have OS vendors tried to brake their OS's intentionally to prevent VMs from working with their stuff. Surely companies writing device drivers aren't going to like guys running their device drivers on top of VMs running on OS's like Linux where one can see "under the hood" and know what is going on with the hardware very easily. Imagine how easily one could reverse engineer one's favourite windows device drivers once device drivers are supported under VMs.

    JUst my $0.02
  • I still fail to see what defines Mozilla's success? The fact that it's open source?
  • All the apps I usually run are easily compiled for pretty much any common CPU. I would much rather have a CPU that ran one ISA lightning fast, rather than one that could run many ISAs but slower.

    Perhaps if you have apps for other architectures that you want to run often the tradeoff would be different, but I am not willing to trade off a slowdown of the common case to speed up a very rare case.
  • No, I disagree. DOSEMU _emulates_ an Operating System. It does not emulate a ix86 machine. One can only run a limited range of dos based programs under DOESMU (I'm told that you might be able to run windows 3.1 in real mode). The point is VMWare actually emulates a MACHINE, providing you with the ability to run the original OS. It's all very well emulating an OS. But OS's such as windows are even more closed source than the standards that modern PCs are supposed to comply with. Why not emulate the machine? that way any new OS will run on it (with possible minor alterations for new technology over time). You don't need to rewrite the entire emulator when win2k comes out for example. Also, all of the broken stuff in windows automagically works without any help. All of a sudden not having a DVD player or support from your favourite hw vendor could become less of a problem for some people. I don't run windows because I don't believe in the methodology and "ideals" behind the beast that is Microsoft - that's not to say others don't use windows. If only freemware were more advanced or VMWare cheaper, Linux could be pre-installed on more machines and those wanting to run their favourite windows game could run it fullscreen inside windows easily without any effort (this is a while away). BTW, emulating a CPU is just plain crazy! The CPU is executing millions and millions of instructions per second, the only times one really emulates CPUs is when designing new ones and one doesn't care how fast they run. Please don't point me to WINE. I know WINE, but it is another OS emulator and not a machine emulator. It cannot ever hope to keep up with windows. Although there are larger and larger gaps between windows releases, there are still new windows releases (at least in the short term) and it takes time for the WINE guys to get cought up (excellent work guys in WINE though - I really find what you do useful, particulalry in my studies when I hjave to ensure that my code will run under the windows IDE that the marker is using - this week it was one of the PROLOG systems for windows to test code written under the standard pl implementation that comes with Linux). In short me need freemware. We kinda needed VMware, but it commercial and we want to have a publically available "free" VM for Linux that doesn't cost more than the OS running on it.
  • First, I don't see that FreeMWare is coming along that great. I'll believe that they are making progress, but I don't see any screenshots.

    Second, VMware is great. It does what it claims. It isn't that expensive. It's a nice piece of software. I have it at work. I don't use it much, but when I do, I'm impressed.

  • Not "raw" SCSI drives, only IDE. Since VMware requires a somewhat beefy system to run acceptably, it makes sense that people who purchase "beefy" systems would be more likely to have fast SCSI drives instead of IDE.

    Err, I highly doubt it. I don't have figured, and I don't think VMware does either (I don't think they're actually tracking this sort of information), but I rather suspect the majority of VMware users have IDE instead of SCSI. I suspect that proportionately speaking, there may be a higher percentage of SCSI users among VMware customers than in the general population, but I'm sure they're still a minority. The average VMware customer is more likely to have Ultra-ATA/66 than SCSI. The kind of "beefiness" VMware requires is lots of memory, but it doesn't require more than the cheapest mainboards available today can provide, so it's not like VMware requires "beefy" systems in the sense that they have to be high-end server-type systems. Just pop a new 128MB DIMM in any typical person's machine and they have all the "beefiness" they need for VMware.


    1. FreeMWare is virtualization, where the processor code is run natively (as just another process; for CPU techies, it's Ring 3) but the VM program modifies it so that Bad Stuff (tm), like executing privilaged instructions and reading stuff that con't be controlled, doesn't cause lots of s..t. Obviously this wouldn't work if the processor wasn't an x86. (Sorry, other replier) Now the different host architectures idea is theoretically possible. Bochs [] is currently working on something like this, where the code is run on a VM, but there is a little mini-virtualization process going on that translates x86 code into native processor code. This can be slow but if you cache the code right you can probably get near-VMWare speeds. I haven't checked how far along this is in development.
    2. Argh! Information overload! Trace to every component in the system!?!? Obviously gotta restrict the devices or the time during which the tracing is performed or both. I think this would also slow things down significantly. What might help is a driver-level trace, i.e., everying that goes into a specific host device from the VM gets logged in some intelligible format. I can't really see how hardware developers could benefit from this, though.
    3. MAJOR STABILITY ISSUE, depending on what you call direct access to a real piece of hardware. If you are going to let some foreign program access the interrupts, shared memory, DMAs, and all that stuff of a device, if the VM crashes it could bring the rest of the system with it. And guess which OS is the primary guest OS for VMWare/Linux, and thus FreeMWare? Exactly. And what do you mean by "host your host O/S? Reboot you VM!"? I don't get it.

    Taling about a "bitchin' development environment": 95, 98, NT, 2000, Linux, *BSD, Be, and a few others, all running in VM's on a Linux host, plus MacOS, WinCE, TI-89 (never mind), etc. in virtualizing emulators. Auto cross-compile through virtualization and wrapper functions that get optimized away. Automatic duplication of test input data with compensation for OS-specific features to test for determinism. Use VNC to remotely control test machines on platforms you don't have VMs for. Auto-tgz or tar.bz2. Automatic snapshot every day, automated upload to web site. Debug automatically picks the host where the rest of the stuff gets in the way least and lets you debug that. Hardware I/O is logged intelligently so that only what you need to see is tracked. I'm sure other people have much better ideas, but that's mine.


  • I've used VMWare, and it does an excellent job of emulating an x86 environment, with better compatibility than Wine, DOSEmu, or just about anything else. That's impressive.

    However, for whatever reason, it needs a lot more RAM.

    You have to understand that DOSemu and Wine are doing different things then VMWare, et. al.

    DOSemu attempts to provide real-mode DOS emulation, with limited protected mode support. There are several things that make this very possible. One is the fact that MS-DOS really isn't much more then a glorified interrupt handler. Another is that the i386 architecture already has support for emulating x86 real mode ("virtual mode"). DOSemu has to setup the processor and service some interrupts, but nothing too outrageous. Remember, DOS programs already have to do almost everything themselves, so there is not as much left for DOSemu to do. (I don't mean to belittle the DOSemu people here; they've done a great job). Memory overhead is low, because (again) DOS doesn't do much to begin with.

    Wine is similar: A project to implement the MS-Windows binary interface and runtime libraries on top of Linux/Unix/Posix/whatever. Same basic idea as DOSemu: Provide the services that an existing Microsoft product already provides. With Windows, though, there are considerably more services to implement. Given Microsoft's love of secret APIs, and the fact that MS has trouble properly implementing their own specification, the Wine team's job is pretty big. Memory overhead is higher then DOSemu, again because it does more. It is still lower then loading all of Windows would be, though, because you already have Linux providing a lot (hardware abstraction, system services, etc.).

    VMWare (and FreeMWare) are doing something very different from Wine and DOSemu. They are emulating an entire i386 machine, complete with protected mode support. The i386 design has very little in the way of i386 virtualization support, so they need to do a lot in software. Furtheremore, they are not trying to run a DOS or Windows program, they are setting up an entire machine. Wine and DOSemu pass as much of the work to the underlying OS as possible. VMware does not -- you have to load an entire OS again. That is why VMware uses more memory: It has to. Because the i386 arcitecture is very well defined, however, they can do a good job of emulating it. Then your Windows program (for example) uses Windows itself to run. VMWare does not need to provide a quality implementation of Windows (if that is even possible); it uses the real thing.

    I hope this sheds a little light on why these programs act the way they do.
  • by osu-neko ( 2604 ) on Sunday December 05, 1999 @03:27PM (#1478830)
    I fail to understand the Linux community's obsession with recreating commercial products - even those commercial products that us. Is there some great reason why we should not pay for VMWare? Is there any single application you people are willing to pay for?

    Recreated commercial products is an excellent idea if there's currently no free alternative. The point is to maximize choice. People should have the option to pay for a commercial solution if they want to, they should not be forced into buying one because they have no alternative. And from the other side, a company throwing serious development money at a team of programmers that can't produce a superior product should go bankrupt! Software is not supposed to be welfare! We shouldn't be paying people to do useless jobs. Commercial developers should make money if and only if they produce products that provide server above and beyond what can be produced by a bunch of hackers in their spare time for free! If they can't, then why are we paying money to them? Is this just corporate welfare?

    Let's get real. Free software is fine and more power to those who make it, but we have to realize at some point that people need to get paid for this stuff.

    Only if they are producing something superior.

    And it doesn't just line someones pockets - its lets them work on their products as a job instead of some "after school" effort.

    If they aren't producing something better than 'some "after school" effort' would produce, they ought not be making money at it! If they are producing something better, they will make money at it, since people who want the better product will pay for it instead of use the inferior free alternative.


  • I completely share your sentiments... As much as the idea of open source seems to be a good one, more often than not it seems more and more open source projects are springing up that simply mimic already created but "proprietary" programs.

    With everyone getting down on MSFT for not innovating enough, where's the innovation in completely recreating someone else's product?

    Perhaps MSFT themselves could take a play from this book and start releasing opensource versions of all of their competitors top products. It wouldn't help them in terms of gaining income from those markets, but at least it would deprive IBM, Oracle, Sun and others of much needed revenues... And they could say with a clear conscience that they've done nothing wrong, since that's just how the open-source community operates.

  • You have to understand that Linux has gone mainstream already. This means that free software developers are no longer the majority of its user base.

    I fail to see how this is relevant to your point. The people who are, as you imply, diluting the number of programmers would never have looked at this project anyway! The only difference is that now they're ignoring it while running a free operating system, instead of ignoring it while running a proprietary one.

  • Perhaps MSFT themselves could take a play from this book and start releasing opensource versions of all of their competitors top products. It wouldn't help them in terms of gaining income from those markets, but at least it would deprive IBM, Oracle, Sun and others of much needed revenues...

    Would it? If IBM, Oracle, and Sun produce superior products, people will pay for them regardless of what Micros~1 does, so they will continue to make money. If Micros~1 releases as open source what is in fact a superior product, anyone, including IBM, Oracle, and Sun can simply incorporate Micros~1's work into their own, and add additional features making their own offering superior and continue to make money. If IBM, Oracle, and Sun decide to respond to this move from Micros~1 by not producing a superior product by one of the two methods mentioned, it becomes highly questionable whether they deserve to make money. Should we just pay them for inferior work just because? I'm not a big fan of corporate welfare, even if it's for a tech company I may work for some day. I'd rather make my money honestly, thank you very much.

    And they could say with a clear conscience that they've done nothing wrong, since that's just how the open-source community operates.

    Indeed. They could say that, because it would be true. If, under the conditions specified, IBM, Oracle, and Sun couldn't continue to make money, it'd be their own fault, not Micros~1's.


  • /* It will swap like crazy if you assign the VM an amount equal to your physical memory.*/

    I don't believe I've used the word 'duh' since the early 1980's.

    I have a K6-2/400 with 256M RAM which makes it very comfortable to run Linux with 128M for host and VM with 128M. Performance ceases to be an issue above 192M RAM -- sometimes VMWare may *APPEAR* to be slower than it is because of screen refresh rates. If you run the VM in full screen mode you can get native speed on a machine with a decent amount of RAM.

  • Don't you mean save up your lunch money for several years so you can afford to buy a SCSI card, then several more years to replace each of your IDE devices that work just fine and without the complications of the SCSI interface?

    SCSI isn't exactly a low cost solution. IDE is.
  • IDE works great for slow CD-Rs... I wouldn't waste a Plextor 32/8/8 on it, but then, Plextor doesn't make IDE CD-Rs (that I know of).

    But, if you spread devices out across IDE channels properly, you're fine. It takes more IRQs, but other than that...

    Four IDE channels is two main HDs on their own, two secondary on a third, and CD devices on the fourth.

    If you do mix CDs and HDs on the same channel, just use common sense. Don't put the burner and the HD with the ISOs on the same channel, etc.

    It's a poor-man's SCSI, but if you have four or less (or eight or less with a board like the BP6/BE6) then it's really just as good.
  • Is there a way to write to the VMWare image file from the host OS? I mean, could you read the files from the Zip with the host, write a VMWare virtual partion with the host, and then real the virtual HD from the guest OS?

    Seems better than trying to do some strange hardware emulation with the parallel port.
  • Open source developers aren't in it for the money. I think most serious(long time) open source developers program on projects becuase they like doing it or to "scratch an itch". Sure appritiation is great, but most developers would be just as happy if their software was simply taken for granted. I know quite a few open source developers and never has it seemed like they were working on what they were working on simply for money or appriciation. many of them have life goals like "get rich, retire early, work full time on open source software", can't you see, these guys are idealists. right now i'm writing a program that demos how to program in a new programming environment(entity, check freshmeat). I'm not doing this for money or any other real motives and i doubt more than 50 people even use it ever. open source for me is just programming for fun and something to do. you could send me donations, cards, mails, cakes, and whatever else but objects like that aren't what drive me.
  • I mean, if Linux is to become just another platform for running proprietary, closed-source software, what the hell is the point? You might as well just run Windows.

    The whole point of Linux is Freedom. You don't get Freedom with proprietary software.

    I will use proprietary software, when it is absoultely necessary. But I will always use a Free alternative, when available.

    I'm starting a new job, and will need VMWare so I can test stuff in Windows, so I have asked my new company to purchase a copy of VMWare. But, I assure you, as soon as I can switch to FreeMWare, I will.

    It's not about the money; it's about the Freedom.

    I would think by now, people would understand that. Even Anonymous Cowards.

  • There will always be a certain ammount of 'fitness' simply because a project is open source, especially if it's GPLed or otherwise locked from becoming proprietary.

    The OSS you know what's in it and know that it won't change in ways that make it less usefull.

    If you're locking into NT, you never know what MS will do. They might 'break' major connectivity preventing some part of your project from working. Many businesses I know use unix for mailservers and other important machines, and NT to admin the lower-end user boxes. What if part of this was broken. You'd have to change how your department ran, or switch to a new OS.

    If you were locked into Linux, this wouldn't happen. There's no incentive for Linus to change protocols to be less compatible with other OSes, and even if he wanted it, he's only the most prominent voice in the primary branch, many distros would simply 'patch' the imcompatibility and continue business as usual.

    Similarly, any closed source software can be modified in ways that are detrimental to you and you can't do anything about it, especially if there are needed patches which are only in the new version. (Think NT and the killer service packs.)

    I do low-end database work for a client who uses Paradox for compatibility with the Corel office suite, and at each new release we have to go through and fix a bunch of problems preventing our old programs from running on the new 'compatible' version. If this was an open project, this wouldn't happen.

    (And yes, I believe you can be 'locked' into Linux, or unix at least (being as replacing unix with BSD or vice versa is fairly easy). Simply have a large number of critical apps working on it without better solutions in another OS. But I don't see it as being a bad OS to be locked into, if you must be locked into something.)
  • VMWare's a small company with one real product (maybe there are others? if so, they're really inconsequential). They had the foresight to think that there would be a market for their product. People found a use for their product. They charge a nominal fee for the use of their product. And their reward? An opensource variant.

    I'm sure the CEO, CFO, as well as all the employees who toiled away there are so excited by this.


    If IBM, Oracle, and Sun produce superior products, people will pay for them regardless of what Micros~1 does, so they will continue to make money.

    The world has already proved that it's not the superior technology that wins, just the cheapest. Witness IDE over SCSI, Win9x over MacOS, WinNT over Unix workstations, Pentium vs. RISC, the list goes on and on.

    It would be crippling to so many companies if Microsoft managed to release an opensource variation of their bread and butter product with even 60% of it's functionality. They (Microsoft) aren't making money from those sales anyways, so if they can prevent others from doing so ("cutting off their air supply") while at the same time adding a new buzzword to their growing repotoire (sp?): an innovative open-source product which embraces developer mindshare...
  • No way. Of course, VMWare still has some issues, a product of this nature will *always* have issues, but VMWare is VERY open and VERY prompt about dealing with problems. Their supports is superb, and the product is DIRT CHEAP, especially considering how well it works.

    I was *AMAZED* at how well it worked, and I went into it very skeptical. And everyone I have shown it too (and I mean people who understand, not peopel who are like 'wow, that's amazing you can run windows under linux!), has been amazed at the performance as well.

    It's great if we are working on an open-source project to do the same thing, but the tone of it should be 'he, let's do our own version! and make it open source!' not 'VMWare is not OSS, therefore, EVIL, and besides, it costs $$$, and it should be free... so let's make our own so we don't have to pay.'

    Really, is it so bad that it is not oss, or that it is not free, when their is nobody else in their market? They built an excellent product, at a VERY fair price, with VERY fair licensing terms. (a larger company would have charged you a fee per-concurrent-VM)....

  • by kijiki ( 16916 ) on Sunday December 05, 1999 @03:12AM (#1478867) Homepage
    VMWare and FreeMWare both virtualize PC hardware. Code executes natively on the host CPU. Playstations, and your intelligent toaster probably do not use the same CPU as your host machine. And non x86 host machines will be unable to run the x86 versions of NT, Linux, or BeOS.

    CPU emulation exists and can do all of the above, however, you don't get something for nothing. CPU emulation (see BOCHS, executor, etc) is incredibly slow in the naive case, and even with complex techniques such as dynamic recompilation, signifigantly slower than native code.
  • by Issue9mm ( 97360 ) on Sunday December 05, 1999 @03:20AM (#1478871)
    I don't mean to be critical, I really don't, and before anyone flaims me, I'm guilty of it too. But here is what I'm hearing:

    "I'd really like to use VMware, but they want money."
    "VMware is great, but they keep nagging me to pay them."

    Etc., etc.. I understand that VMware's not open source, and that maybe it should be. Maybe after this it will be, once they realize that they're not the only kids on the block anymore. That doesn't mean that we shouldn't pay for it. If it's a program that you use, and that you enjoy, you SHOULD pay for it.

    When developers start realizing that they're catering to a bunch of cheap bastards (myself included), they're going to pick up shop, or begin attaching themselves to something a little more worthwhile. Yes, someone else will pick up where they left off, but we need long-timers. Those in for the long haul, who've been around and gotten the experience.

    I for one tend to at least try to support the projects that I reap benefit from. Granted, I don't use VMware, and would probably switch to FreeMWare if I did, but not because it was free. More because I am free, free to decide what I think it's worth to me in the scheme of things, free to choose when or where I'm going to pay for it. In other words, I don't use free software. If I like something, I like to show my appreciation of the effort, and money usually works quite nicely.

    This is not to say that the developers wouldn't be just as happy with a postcard. A lot of times, it's things like that that MAKE the project worthwhile. In short, I strongly urge those of us caught up in Open Source to appreciate the authors. If you're going to switch to FreeMWare, try and make a donation. Can't make a donation? Send a postcard, or an email, or a birthday cake, or something. Let them know that their work is appreciated, or it won't go on much longer.
  • by Elvii ( 428 ) <> on Sunday December 05, 1999 @03:21AM (#1478876) Homepage
    Anyone know what minimum specs for this are? I don't use VMWare because my processer (p233) wouldn't run concurrent os's well... I know I *should* run a beefy machine to do something like this, but for situations like mine, where it's plenty of machine for my use, but not the newest wiz-bang 750 mhz chips, will this work well or at all? Lots of non-cutting edge systems out there, ya know. :)


    bash: ispell: command not found

    bash: ispell: command not found
  • For what you are talking about, Bochs is probably better. FreeMWare, like VMware, is only about virtualization. They make an operating system think it's controlling the CPU entirely. It will only work on x86-based computers. Bochs, on the other hand ( []) is a complete portable x86-emulator. It is MUCH slower, though, cos it has to emulate every command. (VM|FreeM)Ware only emulate the protected instructions (or something like that).

    Anyway, it seems the VMware people have patented some of their technology, so I hope the FreeMWare project doesn't run into any trouble there. I was unable to find out exactly what the patent was.

  • by Anonymous Coward on Sunday December 05, 1999 @03:32AM (#1478885)
    I fail to understand the Linux community's obsession with recreating commercial products - even those commercial products that us. Is there some great reason why we should not pay for VMWare? Is there any single application you people are willing to pay for?

    Let's get real. Free software is fine and more power to those who make it, but we have to realize at some point that people need to get paid for this stuff. And it doesn't just line someones pockets - its lets them work on their products as a job instead of some "after school" effort.

    When the next killer app comes out for Linux I'll pay my fair share. How about you?

  • I can't understand why people seem to have such a problem paying for software that they will (most likely) use to run a commercial OS. How many
    people out there are using VMWare to run NetBSD in a window on thier linux box?

    You've certainly got a point (while it would be fun to run FreeBSD in a window, I'd probably be mostly running Win98 and BeOS on top of Linux). However, there is a very good reason for a virtualization system to be open sourced: easy migration path. People who would otherwise not use Linux b/c they don't want to reboot to use office or play games can use this (which will, I'm sure, be in most distros once it's stable). Despite the fact that they paid who-knows-how-much for windows and office, they will balk at paying $100 to be able to run windows on linux. But if it's open sourced, they will see it as "free beer". Once alternatives appear (which they're starting to), people will already be confortable with Linux and won't have a problem moving to it entirely.

    But overall, I agree with you... if you just want a free-beer VmWare, quit whining and pay the damn company what they're asking! Or help develop FreeMWare. :)
  • Well, the way I see it, there is a difference between "I want it" and "I need it" For example, Quake III is purely "I want it" (except to for those few fragaddicts.) It's not a necessity and it is not going to really make your life any easier, just more fun. So people don't mind paying for it.
    Something like and OS or and Office Program or VMWare becomes more of a necessity. In the case of VMWare it safes lots of rebooting time, which nobody likes to do anyhow. What good is having an OS that can stay up for years just to have to reboot into one that has to reboot practically everytime a application is installed? When something crosses the line of novelty to necessity I think it is more important that we don't have to pay for it.
    Besides if these people want to code this software then let them, I mean people like to code stuff, if this type of thing tickles thier fancy then it isn't just about not paying for VMWare.

  • Okay.
    1) They shouldn't have called it 'FreeMWare'. That's not nice.. it's too close to the name VMWare. It could possibly be open to a (possibly deserved) trademark suit.
    2) It's GOOD that this project exists, however.
    3) It will be quite some time before it matches the performance of VMWare. I'm guessing a year at least, but I'm no expert.
    4) VMWare is a good product. The presence of the (lagging behind) FreeMWare may give them incentive to continue to improve their product over the next few years, or else risk losing marketshare to FreeMWare. This is a good thing, and one of the good benefits of free software, it'll keep the commercial stuff GOOD. Peoples reason for buying VMWare will be, as it is today, that it is the best thing out there, period.
    5) People have to chill. It's cool to start an OSS project about something, but it shouldn't be done to 'spite' a commercial product. Linux doesn't exist for the purpose of replacing window.s. that's just what's happening now...
  • Bochs and VMware are written by the same person. Although Bochs does a similar thing Bochs is a PC emulator that can run on multiple architectures and freemWare is a virtual machine that tries to execute as much code natively as possible. Therefore Freemware will run quicker than Bochs but will only run on x86.
  • by Anonymous Coward

    VMware is definately worth the $100/USD becase it is available and works well.

    Like the Wine project, I'll be looking at the FreeMWare project because it is open source and I can thus trust that nothing odd is happening in the code behind my back. As we've painfully been made aware, any closed source product has a high chance of having some kind of trojan used to send details about us back to the maker. Who knows what other things it might be doing? This is my only real concern with VMware or using other closed source products.

    I think part of the negitive reaction to paying is that it's an admission of defeat; in thier heart-of-hearts they're saying "I have to use some Windows software, so paying for it is a real hit to the ego". (This doesn't count the folks that are students or are just being cheap.)

    Well, VMware is here. You want now, you pay now. You want later, it might be free (as in beer) as well as Freedom. How many of us have tried Wine, and found it's just not usable for what we need? (raises hand) I bet I'm not the only one.

    VMware is usable -- now, not later -- for running FreeBSD or NT, so if you want to learn about these other operating systems without commiting to them, VMware is definately worth the price they ask. It works well, and isn't too hard to configure. If you've compiled a Linux kernel, it's trivial.

  • by JohnG ( 93975 ) on Sunday December 05, 1999 @03:52AM (#1478898)
    One of the reasons I didn't use VMWare is that you have to install Windows ontop of VMWare. That is to say if you already have Windows installed you will have to reinstall and since I have a Compaq I don't a Windows 98 CD and don't really care enough about the issue to ask Compaq for one. (And now if MS gets their way it might not matter anyhow) I haven't used my Windows partition in ages just because I hate having to reboot and there is nothing there that I can't live without.
    So what I want to know is if anyone can tell me if FreeMWare will be able to just boot from my existing Windows partition, or if that is even possible?

  • None taken. :) And I'm down to 5 IDE devices now because the case I have doesn't have any more drive bays.

    I've been round and round with people in this area about SCSI. If I were running a server, it would be SCSI for the speed/reliability/ease of installation. But the home PC [] is the one with all the drives, it plays games, it reads Slashdot, plays MP3's, and has only 'coastered' ~10 out of 200 CD's I've burned.

    In responce to WNight, ABIT BH6, primary disabled, secondary has the CD devices, the Promise Ultra card primary has C: and secondary has D: and E:

    For Rendus, That's my biggest point, I'm not doing anything critical with the system, why spring for the extra? However, I find it rather easy to work with SCSI on the servers at work, just make sure the ID is different and the last drive is terminated and it works great, both the Sparc5 and the sparc10 on my desk run SCSI, and share an NFS mounted external drive and a CD-ROM, but that is at work, IDE is at home :)

  • by JohnG ( 93975 ) on Sunday December 05, 1999 @04:16AM (#1478918)
    Ok I seem to be reading every other post saying "leave VMWare alone if they want to charge money let them how dow you hurt their business!" And the other half of the posts say "Yeah, FreeMWare now we don't have to pay those losers VMWare $100"
    Well, let me just point out that there are companies producing commercial X-servers and doing quite well at it, even thought XFree86 exists. AND not only does XFree86 exist but it comes with every distribution. If these X-server companies can compete with the out-of-box solution than VMWare can compete with FreeMWare. I mean Wine hasn't replaced Windows yet has it? The folks at VMWare just have to raise the bar alittle. Since they are making money they just need to sink a little of it back into the software. If they can't raise the bar and compete with the free stuff then they never deserved to be in business in the first place.

  • Alright, let me be a little more clear. You're 100% right about what DOSEmu and Wine do (although the DOSEmu people haven't released anything lately, it looks like if they do, they'll be looking into trying to trap more of the protected mode stuff, which is basically just what VMWare does too)

    My problem with the RAM is this: 8MB of RAM is perfectly acceptable overhead for VMWare, but the extra 32MB (or whatever) for the virtual machine is not. It should be able to dynamically allocate that RAM, and at least return a page of 0's for any new memory that the virtual machine wants. Otherwise, it should run faster, not both. That is my problem with the RAM.

    VMWare *does* *not* emulate an entire i386 machine. (Bochs and SoftWindows do, and they're much slower, for obvious reasons) If it did, they'd release it for Solaris, and make a buttload of money. VMWare *requires* an x86 machine, because it passes through a lot of the x86 assembler instructions. (wherever possible, trapping and emulating protected mode and other nasty stuff the Host OS won't let it mess with)

    I don't have a problem with that, I just have a problem with the memory. And another poster commented that they make a "flat" address space that to make it easier for them. Well, that's great, and if it's actually impossible to do it well any other way, I guess we'll have to live with that. But it is an issue, and I'm not convinced that it has to be that way, yet. (I guess they'd try to do it the other way if possible. I'm sure they had their reasons, but I'll spend my time rooting for someone to implement it in DOSEmu or even FreeMWare...)

    Other stuff to mention about DOSEmu and Wine:

    DOSEmu has pretty video support and some initial sound support, which is cool. If Bochs was also GPL (I don't think that'll happen, but...) the two projects could merge / share code, and DOSEmu could fix their video and maybe add a little sound, while Bochs could emulate all that DOSEmu lacks, and make something fast that uses less RAM than VMWare. At least, that would be the goal. I guess DOSEmu could share code with FreeMWare, but I haven't checked the licensing.

    Wine is just like the old WABI project, except it supports a lot of new stuff. However, currently it has moved away from getting the old Win 3.1 features perfect--which is okay--and focusing more on the current new features that people need for their personal "killer apps", which makes some good sense (they can't do it *all*, but it's great to see help from Corel). Also, Wine doesn't need an existing Windows installation. In that respect, they've done a lot of work, *more* than VMWare, since they've literally had to reimplement the Windows API. You said that, but I argue that that's *much* more work than even reimplementing the x86 at a low level, IMO, just due to its sheer size. (and they did some of that, too, there's interrupt code and minimal console output in Wine. (for instance, it runs PKZIP, because it has to. :)

    I think that both Wine and DOSEmu have code from the TWIN project to completely emulate an x86, and they're both attempting to merge that in for cross-platform support. But we'll see if anything happens with that.

    But yes, these are all 'emulation' or reimplementations in some way or another, and they are different. But they all accomplish many of the same goals, and therefore I thought it appropriate to make a comparison. Thank you for explicitly pointing out the major differences, though, you saved me some commenting. :)
    pb Reply or e-mail rather than vaguely moderate [].
  • I agree. Right now each OS provide is solving the same boring problems: getting 3d to work, getting driver support for exotic hardware, making a filesystem, etc.

    Having a generic OS core that is free and allows other OS to coexist would mean a great deal for smaller operating systems like BeOS. They wouldn't have to worry so much about hardware support, they wouldn't have to worry so much about coexistence with other operating system and could focus on getting things to work with the virtual hardware instead.

    What I'm curious at is whether a multimedia beast like BeOS can be run efficiently virtually or that it would have to live with the limitations of the host OS (filesystem size, etc.).
  • Besides if these people want to code this software then let them

    Far from me to disagree with that but, on the other hand, think how many really new things we could be doing if we wouldn't be reinventing every single commercial wheel out there.

    Or even some non-commercial wheels: I've seen somebody from Debian suggest that they should make a free implementation of a 2000-line program I once wrote and put on the net for free (back when the net was still a small and cosy place and the GPL had just only been born). Why? Simply because, for reasons beyond my control, I had to disallow "commercial use". Fortunately I found out (by accident) and was able to convince them that they were being worried about nothing worthwhile and should spend their time doing more useful things. Especially since there already is an in part similar GPL-ed program out there as well (by none less that Jamie Zawinsky, himself even).


  • As you might have read in my other post I haven't booted into Windows for many months now. So VMWare isn't remotely important to MY way of computing. Some people however might not want to give up the games of Windows or MS Office. But that doesn't mean they should always have to either use a crummy OS or reboot between OS's two or three times a day either.

  • So you've looked for trojans before and found them? If that is true, I'm not sure that most people, even programmers, would know what a trojan looked like in a large piece of code. I know that there are functions which I've seen which served no apparent purpose. The general case when dealing with other peoples' code is, if you don't understand something, leave it alone, which I'm pretty sure applies to most of the kernel for most people.
  • So you've looked for trojans before and found them? If that is true, I'm not sure that most people, even programmers, would know what a trojan looked like in a large piece of code. I know that there are functions which I've seen which served no apparent purpose. The general case when dealing with other peoples' code is, if you don't understand something, leave it alone, which I'm pretty sure applies to most of the kernel for most people.
  • Yes, you should be able to - as long as it's on an IDE drive and you don't mind rebooting your virtual machine a bunch of times to replace the hardware drivers.

    I can tell you you'll have problems trying to take an NT install on a dual CPU system and make it run in VMware, because VMware only "presents" one CPU to NT when it's running under Linux.

    For 98' you should be able to boot of the "raw" IDE partition. You will in all likelihood get a bunch of "error" messages about missing hardware or new hardware. Going into safe mode and "deleting" all your hardware, including motherboard "resources" and rebooting should make Windows "rediscover" the new hardware under the VMware environment. This is necessary because VMware "substitutes" virtual hardware for some "real" hardware. For instance, no matter what kind of Ethernet adapter you have VMware presents an AMD PCnet Ethernet Adapter to Windows. All I can say is that it worked for me, but depending on your actual hardware you may have "issues."
  • Anonymous Coward wrote:

    I fail to understand the Linux community's obsession with recreating commercial products

    First off, it's recreating Free versions of proprietary software, the fact that the proprietary software is commercial is incidental, there's plenty of Free commercial software out there. Secondly, why they're recreated is simple, we want to be able perform the task the program is designed for, but with all the Freedom, control, portability, etc. benefits of Free software.

    VMWare will never run on Be or an Alpha linux box, FreeMWare might be ported there. A native port of FreeMWare to FreeBSD or HURD might well be faster and more reliable than VMWare run as a compatibility layer. A security or bug fix can be made in FreeMWare without the intervention of any company. FreeMWare can never become unmaintained due to bankrupcy or merger. There's plenty of reasons to work on a Free version.

    Is there some great reason why we should not pay for VMWare?

    No, is there some great reason Anonymous Cowards keep ignoring the fact that Freedom has nothing to do with price? Nobody is stopping you, or anyone else, from paying for VMWare.

    Is there any single application you people are willing to pay for?

    Yes, in fact many people working on Free software development (particularly Free OS development) have purchased VMWare. That doesn't change the fact that they'd rather be using Free software for the same thing, and many people are willing to work towards making that a reality.

    Let's get real. Free software is fine and more power to those who make it, but we have to realize at some point that people need to get paid for this stuff.

    Let's get real. Shrinkwrap software (i.e. proprietary software sold by piece, such as VMWare, Microsoft Office, etc) employs the minority of programmers, but gets the majority of attention. If all shrinkwrap proprietary software companies were to simultaneously go belly up and die, very few professional programmers would be out of a job. Most of us do consulting or custom work. Free software does nothing but help us, it creates jobs and opportunities in the more professional programming positions.

  • What you want is called Rawdisk support. It is well documented by Vmware on their website and often discussed in their newsgroups. Take a look at the following sites for the specifics on how to set it up. I have used Rawdisks from the start since back in April when the beta's were released. ces.html []

    news:// []

  • I have no intention to flame any particular existing project, and thus will not get flamed back for doing so.

    Notice that I was speaking in general terms, and did not at any point express an evaluation of the usefullness of the FreeMWare effort. Hence, feel free to moderate my origonal post down for being off-topic if you want, but at least read and understand what it says first. Now you can flame me.


  • by wct ( 45593 ) on Sunday December 05, 1999 @05:35AM (#1478974) Homepage
    To add to the VMware vs Freemware debate, here's an interesting quote from Keith Lawton, Freemware founder and bochs developer (from an interview [] on :

    I take exception to people thinking that FreeMWare is riding on the backs of VMWare for two reasons. The first being that the Bochs team has talked about this well before VMWare formed their company. The second is that long ago, I received a request from people at Stanford to use Bochs for free for "educational" use. Given that I like to help out educational causes, I of course obliged. Check out where VMWare got its start. Enough said.

    Without knowing the circumstances and amount of truth behind that statement I can't really comment further. What I can say is, I have used bochs before (for an OS design assignment) and while slow and difficult to configure, it is quite versatile and usable - Windows 3.1 runs usably under Alpha, for example. But I wonder how much of bochs is directly applicable to the problem of virtualisation?


  • I was told by the guy from vmware at ALS that BeOS does not run well within vmware. It simply likes to have more resources than can be alotted to it by the host OS. I'm also sure that vmware is better optimized to handle things like windows and linux/bsd due to the much larger number of users for those os's than for the BeOS. I have Be, but not vmware so I've never tried it.
  • When I tried the trial version of a relatively recent VMware a couple months ago, it recognized a pre-existing Windows partition just fine, with the caveat that it had to redetect all the system hardware under Linux, which took a while. I ended up removing it because it ran too slowly and I didn't have the time, effort, or skill to spare to hack around with it and figure out how to find my VMware arse with both hands, so to speak. Maybe FreeMWare would be different. I dunno; maybe I'll fool around with it once it gets finished someday.

    I can't help but think, however, that this sort of thing casts a bad light upon the OSS/Free Software community. Some entrepreneur out there has a great commercial idea, and what's the first thing the OSS/FS people do? They rip it off! Wasn't this precisely the sort of thing patents were intended to prevent? What happens if the VMWare people have/get a patent on their program?
  • Lemme look. Yep, here it is. []

    Don't go bashing them too much, that was pretty easy to find with the "Search" function. Usually they try to put related links in that extra box at the top, next to the story. CmdrTaco's take on reposting old stories (by mistake, this one wasn't a mistake, just an issue that Roblimo thought should be discussed again, maybe too soon...) is that there are too many stories and submissions to wade through (His estimate in Thoughts From The Furnace [] was around 9000)

    However, if a simple search function to find old articles about the (exact) same topic before posting a new one was implemented correctly, it would be very nice. It would eliminate all of the "Didn't we already see this on Slashdot" posts, as you were saying, it could add a link to the archived version going to the new article if needed, and definitely add a link to the new article back to the old one, to let people know that we are discussing this again for a reason. And it would eliminate all posts like your own, because the problem would be solved, the new feature added, and everyone would be happy. Except for the dude who had to implement it (but he'll probably be inordinately proud of it once he's done, so that's okay). :)
    pb Reply or e-mail rather than vaguely moderate [].
  • Some of us may seem cheap for not wanting to pay for software, or pirating it, or whatever. Some of the people who use free software could just as easily use commercial software, and pay for it. However, some of us can't. Some of us have always been in a low-budget family or on a low budget ourselves and would be insane to pay mass amounts of money on software. For some of us, computer hardware itself is received second-hand. What makes people think that we can pay for commercial software!? Those of us who simply CAN NOT feasibly pay for commercial software find ourselves in a paradise with free software. Free for download, at least. The freedoms that Open Source licenses give us are a nice addition. But the money factor is a bigger one. Some people might say that people without much money shouldn't be using computers due to their high cost. Well, the Internet has certainly changed that. Computers can be a way to get out of debt, as eBay has taught us. Collapsing the cost factor for ownership of computers is what people on low budgets need most for surviving in the Information Age. The ideals of Richard Stallman on anti-proprietary software are like the sermons of priests to some of us. Bear this in mind before you complain about software developers not being paid. There still are good souls out there willing to help more and more people become part of the Information Age.
  • by alkali ( 28338 ) on Sunday December 05, 1999 @05:51AM (#1478995)
    While I don't have any idea whether it's technically a trademark violation, I wish the developers had chosen a name that wasn't simply a variation on VMWare. "FreeMWare" is currently a misleading name in that it suggests that it can do everything that VMWare can do -- which I understand is not presently the case. And even if it could, the name itself suggests an unnecessary hostility to the existence of a commercial product. Names are free; there's no reason not to pick something original.
  • by Lemmy Caution ( 8378 ) on Sunday December 05, 1999 @06:03AM (#1478999) Homepage
    This sort of development is a line of demarcation between those who like Linux because, like the BSDs and Hurd, it's Free and Open, and those who just like Linux as the "alternative operating system of the day," like OS/2 and BeOS and AmigaOS were/are.

    I'm a member of the former; I'm glad that there's a Free VMWare-like solution. I'm not so religious that I would never buy commercial software - I do and will - but I will always prefer a Free option, even if paying for media and documentation (money isn't a big issue for me.)

    This DOES put Linux ISVs in an awkward position, but I'm afraid that's really their problem - I hope to see the day that the idea of paying for software is as archaic as the idea of paying for buggy whips. I'm not doing this to make ISVs rich.

    I have a concern, in fact, about the growing success of Linux. If two people are doing something for Free because they enjoy doing it, they will usually work pretty hard and do a good job of it. If both of them are getting paid well for it, nothing changes except - perhaps - things might happen more quickly. BUT if ONE is getting paid and the other isn't, I suspect that the latter *might* say "screw this, I'm getting out of here." I'm nervous about what the move to funded development by groups like Mozilla, VALinux, and RedHat might do to the people who were developing on their own dime - and when the IPOs pay off and we get our first cash-in-hand Free Software multimillionaires, how will that affect the people who *aren't?*
  • Hi guys. I just wanted to put my few cents in. There's a 'mac' version for ppc machines that can run Mac Os. Well, you could run any other os you like. It's called mol ( It's open sourced, and works on just about every mac machine. Anyway, just thought I'd let you know :P
  • Heh. What's a "freem" anyway? Netware, software, hardware, wetware, vaporware, trialware, shareware, freeware, and now this mysterious "freemware."

    Tell me -- it is made with real freems?

  • Hmmm, my other activity got postponed, so I seem to have time to answer this more in-depth...

    There are many elaborate and/or fancy ways to describe how open source works, but in the end they all come doen to this: survival of the fittest.

    However, the important thing is to understand what it means to be fit or unfit. Being (un)fit is something one (or something) does for a certain purpose and within a certain context. Part of being fit for survival as an open source programmer is to recognise when a certain area of interest has already been occupied by projects that are fit and established enough so that fighting them is likely to be a loosing battle.

    Note: While M$ Windows is very well established, it is in general not sufficiently fit. If not from technical point of view, then at least from the point of view of matching what people in certain "markets" want from a psychological point of view. That's why it makes sense for open source projects to enter the PC operating system arena. But this is an evaluation that must be made over and over again for each new (open source or other) project, and it is not always so that the new project will win, even if it is "better" at something (for some definition of better). Certainly not if the all established projects in the given arena already are open as well, as is the case with the example in my first post in this thread.

    By the way: an other frequently misunderstood but crucial aspect of survival of the fittest is that it are not always the fittest who survive. The fittest survive on average.


  • by pb ( 1020 ) on Sunday December 05, 1999 @06:27AM (#1479015)
    I've used VMWare, and it does an excellent job of emulating an x86 environment, with better compatibility than Wine, DOSEmu, or just about anything else. That's impressive.

    However, for whatever reason, it needs a lot more RAM. It has to physically allocate however much RAM you tell it to use for the emulated OS, in my case 32MB for Win '98, and then it uses at least another 8MB for its devices and itself, and somewhere in there my 64MB K6/300 decides that it hates life and gets really slow... That's why they recommend at least 128MB RAM. DOSEmu, by contrast, never uses as much RAM as I tell it it can use unless it absolutely has to. Usually I give it 8MB, but when I wanted to run Callus, I gave it 20MB. Worked great, except for lacking sound. Wine generally uses 4MB above and beyond the memory usage of the Windows app, in my experience. (these numbers are all pretty rough, if you've tested this more, please post some results)

    Also, I didn't like it that VMWare didn't support more options for an x86 drive. I have a lot of ext2 partitions that I use for my DOS stuff, and DOSEmu and Wine deal with that just fine. I guess I could make some native FAT partitions, but those things are nasty. And compressed drives really are a hack, but I might do that again instead. So I've got a big file where VMWare keeps its 'OS'.

    And, when all is said and done, what good is it? Well, I've found that I don't really have much of a use for Win '98, and I can run a lot of other stuff with DOSEmu or Wine. Just about the only thing I'd want VMWare for would be displaying videos with proprietary, unsupported codecs, since XAnim is missing a lot of them and the companies are pretty lame about it.

    So why would I want FreeMWare? Well, to play around with it. To be able to compile it with my compiler optimizations and see how it runs. (even if the VMWare team does something like this... well, I don't know about it, and I can't test it)

    To see if someone hacks in ext2 support or some kind of generic drive emulation that works well. (have the IDE/SCSI faking area, or use Linux's SCSI faking, and then have the actual drive, whether it's a disk file, a FAT partition, a DOSEmu drive, a VMWare drive, or an ext2 partition...)

    I'd like to see it without the weird video corruption I get with VMWare (although my video card does suck :).

    And then I'll have to test out how the native sound works in DOS, that's a must for my DOS games. And then benchmark against DOSEmu. :)

    Of course, first I'd like to know how it's doing now. Has anyone built the source from CVS? I normally just download the releases, but the warning on this one indicated it was anything but stable.
    pb Reply or e-mail rather than vaguely moderate [].
  • Go get Emacs and CVS (both available for BeOS, your proprietary operating system). Start writing code for the BeOS FE.

  • I'm curious, there's Open Source software movement, why isn't there an Open Source hardware movement. It would be harder, but possible I think.
  • What I hear you saying is that you like things which are open source because you "know" they don't have a trojan, or at least that you can check. However, I ask you whether you've every really checked the complete source to an reasonably large open source project? Do you really think you could read through FreeMWare and understand it well enough to know that there wasn't a "trojan"? I doubt it.
  • An idea would be to bootstrap vmware or freemware ontop of an exokernel (I just read the HURD story). The idea of an exokernel is to put very little functionality in it and focus on multiplexing the hardware (so that it can be used by multiple applications). Using such a setup would make it possible to host multiple normal operating systems without having to use a limiting host OS.

    I really like this idea of being able to run multiple operating systems at the same time since there is no one size fits all OS. Different applications come with different requirements. The ability to serve web pages is something entirely different than the ability to handle large files.
  • I've run it on a P3 500, but it still lags if you don't have enough memory and it starts swapping. I think that 256 MB is ideal. I think if the other people that replied had more memory their performance would have been a lot better since besides going in swap for memory, context switches between the real and the virtual machine seem to happen very quickly and efficiently.
  • I think we should all donate a bunch of money to the FSF or other free software project. We users can do our parts.
  • I have a pentium II 350 and VMware runs OK. I have 128MB memory, and I let the VM have 64 of that. VMware is quite a pig, and it uses a small slab of memory *on top* of the memory that you assign. It will swap like crazy if you assign the VM an amount equal to your physical memory. Given this, and the fact that I wouldn't recommend any OS on less than 32MB, you probably want at least 64MB of memory to run VMWare.

    I don't think FreeMWare will be much better at this stage.

As of next Tuesday, C will be flushed in favor of COBOL. Please update your programs.