Do you know when boinc will utilitize the 64-bit processor?

---

When life is to big to tackle, kick it in the shins.

It will run fine on a 64-bit processor. I know that isn\'t what you mean but a 64-bit version will not provide much advantage to any of the current projects. The current projects are floating point heavy and the FPU on 64-bit processors is the same as the one on 32-bit processors. If a project was integer heavy or needed a large address space then a 64-bit version would have more impact.

---

BOINC WIKI

BOINCing since 2002/12/8

I\'d like to mention that I just built Boinc for my opteron workstation. It works just fine, except I don\'t think it can do anything, because the server doesn\'t know what application to send it... Here\'s the message:
[BURP] Message from server: platform \'x86_64-unknown-linux-gnu\' not found

Of course, I shouldn\'t have that -unknown- in the middle of it, but I\'m working on it. Not too sure where to change that though.
Anyhow, I\'ll probably just download the 32 bit version of boinc anyway.

---

What\'s the gain for your effort? I can\'t see how an optimized core client would help anybody since it only manages your projects and doesn\'t do any computation by itself. That\'s the same as with SSE & Co. optimzed core clients - they are only used to have the benchmark corrected if you run a manually optimized project client (currently only available for SETI).

---

And currently it doesn\'t seem like there\'s a 64bit version of Blender on the way quite yet...

Blender has always been 64 bit (since the NaN days). You just have to build it on a 64 bit system.

---

Blender has always been 64 bit (since the NaN days). You just have to build it on a 64 bit system.

Yes but does it utilize the extra 32 bit for anything?

Good question- not having used a 64bit build, I couldn\'t say, other than that there have been some errors noted in taking a file created on 64bit blender and opening it in 32bit blender (Ie it doesn\'t open), which suggests to me that there are significant differences in the variable structure between the two.

I\'m going to ask this question on the bf-committers mailing list, (which btw is a pretty valuable resource if you haven\'t come across it Janus), and I\'ll post the answer back here

---

Edit: I found this thread on the archives:

http://projects.blender.org/pipermail/bf-committers/2006-March/013816.html

Maybe that is the answer?

---

Also:

http://www.blendernation.com/2006/03/07/64-or-32-bits-that-is-the-question/

It seems that the only major advantage of 64bit is speed?

---

Yup, and that\'s not really because of the 64 bit but rather the extra on-chip memory registers that the AMD 64bit core systems have.

Theoretically there should be nothing wrong in compiling a 64bit version of BURP+Blender - in theory it should work out of the box. It\'s just not the time right now. Stability first, optimizations later.

I am running Debian Unstable on AMD64, i installed boinc from the repository but how do i get burp working?

---

So far, three projects have added support for AMD64: SETI sends the x86 application and SIMAP and Chess960, native x86-64 applications. Would you consider supporting windows_amd64 and x86_64-pc-linux-gnu/x86_64-unknown-linux-gnu similarly?

Thanks.

---

So far, three projects have added support for AMD64: SETI sends the x86 application and SIMAP and Chess960, native x86-64 applications. Would you consider supporting windows_amd64 and x86_64-pc-linux-gnu/x86_64-unknown-linux-gnu similarly?

HashClash now supports AMD64 with a 32-bit Linux application...

---

So far, three projects have added support for AMD64: SETI sends the x86 application and SIMAP and Chess960, native x86-64 applications. Would you consider supporting windows_amd64 and x86_64-pc-linux-gnu/x86_64-unknown-linux-gnu similarly?

HashClash now supports AMD64 with a 32-bit Linux application...

HashClash now supports AMD64 on Windows with a 32-bit application...

---

FWIW, running two instances of the client, one the 32-bit client, the other, the 64-bit client, on the same 4-core system, but limiting each client to 2 cores, I can compare the relative performance of 32-bit and 64-bit SIMAP\'s HMMER: the 64-bit version is about 7% faster.

By enabling vectorization (supported by default on AMD64), the SIMAP developers observed other 8% improvement.

That\'s a potential performance improvement of 15% when porting the project application to AMD64.

---

So far, three projects have added support for AMD64: SETI sends the x86 application and SIMAP and Chess960, native x86-64 applications. Would you consider supporting windows_amd64 and x86_64-pc-linux-gnu/x86_64-unknown-linux-gnu similarly?

HashClash now supports AMD64 with a 32-bit Linux application...

HashClash now supports AMD64 on Windows with a 32-bit application...

Guess what? Leiden Classical now supports AMD64 on Linux with a 32-bit application. ;-)

---

You shouldn\'t try to convince me - the BURP part of our client supports 64bit pretty well, however none of our renderers (at least to my knowledge) have any official support for 64 bit yet.

I think that what he\'s suggesting is to distribute a 32bit renderer for AMD64 and X86_64 clients; that way it wouldn\'t take advantage of 64 bit processors, but it would allow 64 bit computers to at least run burp? Sort of a workaround before the renderer\'s officially support 64 bit?

---

Just want to put in my vote for a 32-bit binary so 64-bit platforms can be supported without having to resort to a chroot.

---