Xeon or opteron? [revised]

[orbitalpunk]

Hi people,

I am building a new modeling computer and am going crazy weither i should get a dual opteron or dual xeon system. My budget is about $350 per cpu so it looks like it'd would be between a AMD Operon 144 or a Intel Xeon 3.0G 800FSB based on the Nocona core.

My other concern is what would be best for the future of 64-bit computing and what would benefit most when maya is released in 64-bit form running on windows XP 64-bit version. Is Intel's EMT64 really 64 bit computing? it seem AMD64 is in the Opterons is. but am still confused about Intels EMT64.

Any opinions would be greatly appreciated. This is for Maya 6.

Thanks

[orbitalpunk]

geez, i think this new review answers my question.

http://www.gamepc.com/labs/view\_content.as...eron&page=1

Maya benchmark is on page 11. Looks like Opterons are it. I kinda was rooting for Intel tho. But overall, i want what renders fastest in Maya and if Opterons are the ones, that thats where im headed.

[joojaa]

Definetly opteron... besides sp 2 gave the opterons a nice boost so...

Invest in a good motherboard tough, as its the most important component speed and stability wise. I suggest tyan, memyself I have a tyan tiger as my home workstation with two 246 pocs as a cheap home workstation it goes well, havent had any porblems with it after initial bios update day one.

Offcourse at work i have slightly better workhorse, mainly better grephics card and slightly more memory.

[mag]

Absolutely opteron.. cheap and fast

[orbitalpunk]

hey , thanks for replying. I was looking at the tyan as well. but i am debating over the tiger and thunder. the thing about the thunder (2885) is that is states it has a AGP 8X / 8X pro slot. when i enlarge the photo, the agp slot looks different then a standard agp slot so i asume thats a pro slot. my question is, can a agp pro slot accapt and standard agp 8x video card? I have a Nvidia Quadro FX500 AGP 8x card and would like to know if it will fit into this slot. here is a link to the boards page at tyan

http://www.tyan.com/products/html/thunderk8w.html

you can also download a larger image here.

ftp://ftp.tyan.com/img_mobo/i_s2885.tif (1.7MB)

Thanks for any input

[user_deleted]

Am I the only one a little confused with the Maya Zoo Render benchmark results? Looking at the results the dual Xeon EMT64 with Hyperthreading enabled performed slower than with it disabled though on 3DSMax render benchmark the dual Xeon over all performed better against the dual AMD Opteron. It's possible the test results suffered due to using a 32-bit OS instead of a 64-bit OS that could take advantage of the 64-bit architecture of both processors. This is in reference to GamePC quote "As for 64-bit performance, we can?t really tell anything at this time. While we were able to load our latest build of Windows XP 64-bit Edition on both systems and were able to run some quick 64-bit SiSoft Sandra benchmarks, these items combined are not enough to get a good picture of the 64-bit performance of these chips. We?re planning on re-visiting these processors in a future lab report with 64-bit Windows XP, Server 2003, and Linux, for a deeper look at performance and compatibility between these two processors."

[orbitalpunk]

yes, i have seen this issue come up a lot with Hthreading. It seems max likes it, maya doesnt.

check out these comparisons from toms hardware. the first page with max, intels w/ HT dominate.. but in the second test with maya intels perform better without HT and also opterons dominate. opterons dont use Hthreading.

also another factor i have recentlyt learnied is this. if you use max, use xeons, if you use maya, use opterons. why? because max uses SSE3, maya does not. and xeons have SSE3, opterons dont. but from what i've read, opterons will have SSE3 in there next revision. When is still unclear tho.

[user_deleted]

These other links may help clarify things: http://anandtech.com/linux/showdoc.aspx?i=2114 , http://www.thejemreport.com/modules.php?op...cle&sid=126 (, http://www.pcstats.com/articleview.cfm?articleID=1665

[orbitalpunk]

question,

would memory with a CAS Latecy of 2 make a big difference from 2.5? it sure does on the wallet. how about performance wise. Rendering especially.

thanks again

[user_deleted]

It really depends on the RAM. I use OCZ performance RAM that allows overclocking but I stick with the recommended settings from the manufacturer. If you know who manufacturers your RAM (displayed on each RAM stick) then go to the manufacturers website for a listing a recommended motherboard settings (ie: CAS, Voltage). There may also be default settings like 2-2-4-2 indicated on your RAM stick to assist you. I've learned from experience it's best to follow the manufacturers recommendations then to tweak hardware that may get fried.

[orbitalpunk]

h5 ya,

i think you misunderstood the questions. im not trying to overlcock or up the latency on my memory thru the bios. im actaully gonna buy CAS 2 memeory sticks. thats there factory speed. no tweaking. but there about 200 dollars more in price. I wondered if it was worth it. or should i just buy memory with a factory speed of 2.5.

thanks agian tho

[gmask]

rendering is ram and CPU intensive so I would expect a slight speed increase with faster ram and over a years time it should pay for itself. Maybe you buy the ram from a place that allows returns and buy both type sof ram and do speed tests then return one or the other.

[mag]

i think gamepc had an article not to long ago on cas latency where they found out it makers wayyyyyy less difference than it used to .. like a 1.5% difference between cas2 and 3 in photoshop. and almost no difference between 2 and 2.5

[1000101-]

In reply to:Am I the only one a little confused with the Maya Zoo Render benchmark results? Looking at the results the dual Xeon EMT64 with Hyperthreading enabled performed slower than with it disabled though on 3DSMax render benchmark the dual Xeon over all performed better against the dual AMD Opteron.

The way hyperthreading works it is entirely possible that maya's renderer doesn't properly split the load accross the logical CPU's resulting in slower performance than with hyperthreading off.

Max I believe has made some hyperthreading specific enhancements. i.e. not throwing the exact same sort of work to every CPU but splitting it in such a fashion that when logical CPU 0 is working on something; logical CPU 1 is working on something else that would otherwise be an idle portion of the CPU.

I hope that makes sense.

[wtfysmf]

Hey guys,

My first post on this forum !

These tests make clear that which processor is faster depends on the software is used.. But isn't it so that both 3DSmax and Maya use the mental ray renderer more often? Aren't there any comparisons between these two processors with the mental ray renderer? Would be far more useful ( at least to me) I think..

[edit] Okay found out just a bit further down there's someone who tested this out himself: http://www.highend3d.com/boards/showflat.p...p;o=&fpart=
Turns out the Xeon is faster after all (at least with Mental Ray)..

[joojaa]

Note teht opterons get a slight boost when installing sp2