Will we boot Win XP by the deadline?

[sud0n1m]

Im curious to know what people are thinking. We dont seem to be too much closer yet to getting Windows XP booted on the Intel Mac, but there could be a lot we dont know since people dont have a huge incentive to share when there is 11,000+ involved. Do you think we are going to boot windows XP by the deadline?

[soisauce001]

My vote was to wait til Vista.... But I bet you there is some kid out there who's this computer genius who can solve the problem.. The only thing is, I bet this kid can't afford an iMac so he can't work on it or test it..

[sud0n1m]

Anyone confident enough they can do it should go out and get an iMac. We're talking 11 grand + here! Plus, if they decide after a week that they cant do it they can always return it.

Here's the math:
The new iMac starts at 1299, add on 10% as an approximate tax in yr area to bring you to 1428.90. Lets assume that the contest prize makes it to 12,000 all said and done. That would be a payoff that is 8.39 times the initial investment. Better odds than most sports bets. Sure, there are other things you could do with $1430.00, but if you were confident in your ability, the 8.39 times payoff would make it worth it. Plus, if you are able to do it and return the mac after, you just made a killing with no investment other than your time.

Now, lets discuss the time involved. Assume that you spend 20 hrs a week on the project between now and the contest end date. That would be 8 hrs on saturday, 8 hrs on sunday and maybe 4 extra during the week. The contest began Jan 23rd and ends on March 23rd. There are 8 weekends between then and now, or 160 hrs spent working on the project. Assuming you win, you would make more than $75 an hour for your time ($66 if you include the initial investment).

Of course nothing is guaranteed, but that is why there is risk involved. Solve the problem, get 12,000 (estimated). Dont solve it, get nothing. The people working on it make that choice, but if you have the skills, there should be no reason not to go out and get a mac to do it.

[soisauce001]

unless you're 11 years old and your only source of income is moms and pops... And moms and pops aren't going to drop 1400+ to buy their 11 year old a iMac.. No Asian parent I know.. at least.. Anyways.. everything I said earlier was all hypothetical.

[sud0n1m]

unless you're 11 years old and your only source of income is moms and pops... And moms and pops aren't going to drop 1400+ to buy their 11 year old a iMac.. No Asian parent I know.. at least.. Anyways.. everything I said earlier was all hypothetical.

How do I moderate +5 funny?? If there are any 11 year old Asians whose parents wont get an iMac for you. Let me know, I might be able to hook you up. if you arent an 11 year old Asian, dont bother applying - I know if you beg hard enough your parents will get you an iMac.

[dantheman82]

Or you could look at it this way...
I'm slightly confident I could do it although it would require a bit of time and tinkering - and there's always the likely possiblity that just when I have figured out the solution, I discover that someone has posted it first and taken the 11 Grand. Furthermore, I for one and probably others don't really want to buy the iMac, especially considering others have tinkered around for some time with it already with no success. Rather, we'd like to wait around for this MacBook Pro laptop which is apparently running behind schedule in case this whole thing doesn't pan out.

Just a thought...and I've been a bit more quiet on the forums and whatever because my initial "successful" solution has basically been shot down due to the EFI combined with dual-core that form a rather formidable roadblock...

Dan

P.S. I vote that it can be done, but will require more time...maybe I'll be surprised though...

Anyone confident enough they can do it should go out and get an iMac. We're talking 11 grand + here! Plus, if they decide after a week that they cant do it they can always return it.

Here's the math:
The new iMac starts at 1299, add on 10% as an approximate tax in yr area to bring you to 1428.90. Lets assume that the contest prize makes it to 12,000 all said and done. That would be a payoff that is 8.39 times the initial investment. Better odds than most sports bets. Sure, there are other things you could do with $1430.00, but if you were confident in your ability, the 8.39 times payoff would make it worth it. Plus, if you are able to do it and return the mac after, you just made a killing with no investment other than your time.

Now, lets discuss the time involved. Assume that you spend 20 hrs a week on the project between now and the contest end date. That would be 8 hrs on saturday, 8 hrs on sunday and maybe 4 extra during the week. The contest began Jan 23rd and ends on March 23rd. There are 8 weekends between then and now, or 160 hrs spent working on the project. Assuming you win, you would make more than $75 an hour for your time ($66 if you include the initial investment).

Of course nothing is guaranteed, but that is why there is risk involved. Solve the problem, get 12,000 (estimated). Dont solve it, get nothing. The people working on it make that choice, but if you have the skills, there should be no reason not to go out and get a mac to do it.

[starvbasic]

I will have everything working by the end of the deadline.... Being a PC person for as long as I have been I know I can get everything working by the deadline.......

[LonV]

I will have everything working by the end of the deadline.... Being a PC person for as long as I have been I know I can get everything working by the deadline.......

I love it! This is the kind of person and attitude that can move mountains! (btw, no sarcasm is intended....I love this stuff)

[noved84]

Well, also, I'm sure the REAL nerds (myself included) are waiting to get their macbooks and never ordered an imac.

[dantheman82]

Well, also, I'm sure the REAL nerds (myself included) are waiting to get their macbooks and never ordered an imac.
Exactly, I just called the store and it's roughly a week from delivery there. I'm not sure what the status is on any of your orders at this point.

Dan

[jordy]

I follow the ongoing progress/news about this site and admire your idea to set it up! So thumbs up and good luck!

here is what i spoted over this morning:
the Inquirer - Why XP will never officially work on a Mac (http://www.theinquirer.net/?article=29710)
Maybe it may be interesting read for all visitors and gives a share on this particular topic!

Note: I Voted that The Contest is doable but not in the time frame given!
Hope I'm wrong and it will be acheaved on time! :D
Do you considered time extension or it make no sense?

[sud0n1m]

From the blog: (http://windows.onmac.net/2006/02/15/fueled-by-antimatter/)

I was very frustrated to read the analysis by Charlie Demerjian of ?Why XP will never officially work on a Mac?. His stance certainly makes for a more interesting article / opinion, however it also is extremely misleading, and may be discouraging to some who have contributed to the project or are currently working on a solution.

One thing the article gets spot on is its criticism of Apple for neither confirming nor denying the ability to run Windows on the Intel Mac. Apple has acted irresponsibly on the issue by making statements that leave so much up to interpretation.

However, the author is convinced that running Windows XP requires some kind of major surgery on the computer making it infeasible in a corporate environment . What I envision will be a reasonable solution for a corporation and could provide wider adoption is to have a Boot CD that loads the solution. It gives you 2 options:

1. Load the ?CSM? (I use this term for convenience)
2. Return the system to factory settings

You should only have to load the ?CSM? once and then you can boot Windows XP, or Linux or whatever else you would like from its install CD (and later from a boot manager). If you need to go back to the old settings, just pop the CD back in and change it back. The other OSes would remain, but you wont be able to boot them. A simple but elegant solution like this should be possible given everything we know, and should be a reasonable extra step if one is competent to install Windows.

If you havent noticed by now, I despise words like never, cant and wont. The people who use those words never, cant and wont do anything in their lives, and will try to convince you that ?never, cant and wont? apply to you too. If you listen to them you ?never, cant and wont? do anything either and they can say ?I told you so?.

We will dual boot Windows XP and OS X on the Intel Mac.

[SiliconAddict]

three point five possibilities IMHO.

1. Someone is waiting until the last minute to post their instructions along with the much needed Windows hacks so they get the largest prize.

2. It may simply take sitting back and waiting. It may happen at a later date.

2.5..the thing is if Vista is found to be more easily installed on the system I expect XP "research" to fall by the wayside as more and more people work on Vista. XP seems to be a sinking ship. There are too many dependencies that just aren't there in Vista or won't be soon enough.

3. Real research will begin after this prize disappears. After thinking this over I'm starting to think this cash pot is doing more harm then good. How many people are hording their tests/experiments in the hopes that they can win the pot? There isn?t as much sharing of knowledge as I expected and I think it comes back to the 12 grand prize.

[bcgrafx]

I found a post that claims to have succeeded in booting Win XP on the Intel Mac. It doesn't use the methods that are forbidden. I wish I had my MacBook Pro to try this out. Waiting on the compnay purchase.
After that, I'll be waiting on the delivery. Oh well. Love the idea behind this site and the fact that it will go to charity. The industry needs positively reinforced drive like this site provides. Congrats!

[sandman]

Hmmm...

Well, IMHO Windows XP can works on mactel.
Mactel are very similar to a normal PC except the use of EFI.
Using an USB storage device, it is now possible to run linux on using elilo. so, IMHO, with a piece of work/research, it will be soon, at least, possible to boot winxp if it is installed on an external USB2 hardrive (lilo project and elilo are close together so if lilo can boot XP on a PC i assume elilo is or will be able to boot XP as well). Well I know getting XP on an external drive ain't the purpose of the Game but it think, in this case, those who have interest to get windows on their mactel (i don't) should follow the linux-on-mactel way.

By the way, i'm sure this goal will be achieve... soon or later.

[clay]

I'm sure it can be done by the deadline, I only hope somebody doesn't beat me to the punch. I didn't read all 1000+ pages of the EFI spec (printed four pages per page) for nothing! :)

I'm waiting for my friend's Macbook Pro to arrive so we can start testing. It was delayed a week and now he's out of town for a week, so I won't get my hands on it until this weekend, but by then I should have enough code written to start making some decent progress over the next few weeks.

I disagree with those who say that the prize is hindering development. Sure, people working on this problem, myself included, don't want to share their ideas because they want to win the prize, but that changes once we reach the deadline. If I haven't solved the problem by then, I'll be happy to share my work with the community. But since I'm dedicating a large chunk of my time each day to this problem, it's nice to know there's a jackpot waiting for me at the end.

[dantheman82]

to my earlier post with respect to the hurdles.

Let me know if any of these have in fact been overcome. And be careful not to fry the computer...I hear they're pretty expensive to buy.

At the outset, running a Knoppix LiveCD or booting a very basic version of Linux with hacked drivers is NOT getting Linux to work on an Intel Mac. You are circumventing the entire boot process with the Knoppix LiveCD as it is a simply a bootable *NIX OS.

That said, the prize is a great idea and for the effort, it is definitely worth it...

[sandman]

hi,

actually knoppix does not boot at all on mac intel.

but i just checked out the linux-on-mactel website and they now explain how to get a dual-boot linux/macOS...

[John_K]

I usually don't post to forums but I think in this case it's warranted. First of all let me clear a few things up. There is more to Windows use of the BIOS besides the bootloader, hence using a Vista EFI bootloader will not just "make it work". In order to make that solution work there would have to be much hacking of Windows system files to remove the BIOS calls from the non-bootloader stuff (refer to the Windows BIOS calls document mentioned in this forum to see what I'm talking about).

There are only two real ways to accomplish running Windows on a Mac, one of which is writing a custom CSM for Apple's EFI implementation which is hindered by the fact that Apple chose to completely disable the memory region that a BIOS usually resides in. If you don't believe me, fire up the EFI Shell and do a listing of memory areas or use mm and try to write to the BIOS region (the area right below 1mb), you will not see that memory area in the listing nor will writing to it succeed.

Now there is one more theoretical way to get Windows running but it requires *much* work and is actually the solution that I am working on now with a few other people and that is making use of a new feature that is built into Intel's new processors. I don't want to go into many details about it but anyone who has been serious about the competition should know what I'm referring to.

Hopefully this dispells some rumours and gives some of you naysayers a little hope. When we have some results, we'll show them to you so don't come begging for things now. Debugging things running inside EFI is a very time consuming and non-convenient process.

[technotes]

...There is more to Windows use of the BIOS besides the bootloader, hence using a Vista EFI bootloader will not just "make it work".
That's correct but even more important, don't assume that the Vista loader will be able to directly load previous versions of NT. Definitely don't assume that an EFI loader will load a version of Windows that uses BIOS.

In order to make that solution work there would have to be much hacking of Windows system files to remove the BIOS calls from the non-bootloader stuff (refer to the Windows BIOS calls document mentioned in this forum to see what I'm talking about).
The critical one is INT 10 used by the video subsystem.

There are only two real ways to accomplish running Windows on a Mac, one of which is writing a custom CSM for Apple's EFI implementation
Wrong! There is no way architecturally to load a CSM into an EFI implementation. The EFI interfaces are clearly insufficient. CSM isn't an EFI module; it's a hack onto the underlying framework. It's a "big switch" that selects either the crufty BIOS API or the new, shiny EFI API. Once that switch gets thrown, there's no going back without a hard reset.

...which is hindered by the fact that Apple chose to completely disable the memory region that a BIOS usually resides in. If you don't believe me, fire up the EFI Shell and do a listing of memory areas or use mm and try to write to the BIOS region (the area right below 1mb), you will not see that memory area in the listing nor will writing to it succeed.
Have you tried changing the MTRR's? Make sure you call exit boot services first.

Now there is one more theoretical way to get Windows running but it requires *much* work and is actually the solution that I am working on now with a few other people and that is making use of a new feature that is built into Intel's new processors. I don't want to go into many details about it but anyone who has been serious about the competition should know what I'm referring to.
You mean the new virtualization support?

There is at least one more way (theoretically) that's a lot easier than virtualization. You load a program that implements the subset of BIOS needed to run XP into the areas of memory normally used to shadow ROM. Minimally, you need to implement video, keyboard, disk, memory and ACPI tables. That should get through NTLDR and into Windows initialization.

But, then you run into the display driver problem. As near as I can tell, the Apple video controllers don't contain the same firmware information needed by their Window's equivalents. Don't expect the Windows drivers to run without at least some tweaking. You can probably run headless for the initial bring-up but eventually you'll want a display. You'll probably need VGA emulation at some point (both the BIOS calls and the frame buffer).

Or, better yet, just wait for MS to make a universal binary of Virtual PC; it'll be a hell of a lot easier :cool:

BTW, this clearly can be done. The issue is whether it's practical for someone to do it without the assistance of Apple or Microsoft. Except for virtualization (which technically isn't dual boot), none of the approaches seems to be supportable (meaning - deployable by corporations).

-tn

[smurfs]

i vote for it
maybe not we, coz i cannot do it :p

[TJCacher]

But, then you run into the display driver problem. As near as I can tell, the Apple video controllers don't contain the same firmware information needed by their Window's equivalents. Don't expect the Windows drivers to run without at least some tweaking. You can probably run headless for the initial bring-up but eventually you'll want a display. You'll probably need VGA emulation at some point (both the BIOS calls and the frame buffer).

Do you (or anyone) know if Vista contains the same dependencies on Int10 BIOS calls? I'm betting it probably does.

For me personally, I'd rather have a stable, full-featured virtualized Windows machine running as a guest OS under OS X. As long as the hosted Windows session capably supports Microsoft's corporate networking standards, it seems to me that this would overall be just as acceptable to most IT departments as a dual boot scenario.

I know there's another contest for that, but I'm surprised that the organizer of this contest didn't include this possibility in his original rules, given his (apparently) mere need to use his new Intel Mac on his company's Windows network.

Not a good solution for the Windows gamers, but everyone else should be happy, n'est pas?

PLM

[technotes]

Do you (or anyone) know if Vista contains the same dependencies on Int10 BIOS calls? I'm betting it probably does.


In: http://www.microsoft.com/whdc/system/platform/firmware/lddm_bios.mspx

...to avoid the stability problems caused by legacy BIOS communication methods, the Windows Vista? display driver model (LDDM) does not support Int10 communications. As an alternative, LDDM display drivers have full access to ACPI methods to control the graphics hardware.

Certainly some of the INT 10 dependencies have been eliminated but it's hard to tell if they all have. :confused: