Xeon or Opteron?

[JDex]

Xeon or Opteron...

#1 suggestion:

Search Button... this is a weekly topic.

[kex]

grrr brain cant take this again

we need sticky

[maninflash]

I have posted some render results that I did comparing xeon and opteron in technical forum, search and you shall find, but your quick answer would be: Dual opteron 250

[Horik]

I wouldn't go with an opteron system. According to Tom's Hardware Guide the Xeon out performed opteron in every 3d benchmark test. If you don't believe me read the article—"Battle of the Titans Xeaon VS Opteron" at www.tomshardware.com (http://www.tomshardware.com) .

[Fahad]

I have posted some render results that I did comparing xeon and opteron in technical forum, search and you shall find, but your quick answer would be: Dual opteron 250


I wouldn't go with an opteron system. According to Tom's Hardware Guide the Xeon out performed opteron in every 3d benchmark test. If you don't believe me read the article—"Battle of the Titans Xeaon VS Opteron" at www.tomshardware.com (http://www.tomshardware.com) .
take a trip with me and see what the future of this thread will look like...

maninflash: no, according to www. (enter uber tech web site here) .com the opteron is (enter figure, speed, etc here), so please (some condesending statement about the other guys knowledge and sources)
Horik: i'm sorry that you think (the condesending statement mentioned above) about me, but (enter come back condesending statement here) because according to www. (enter ultra uber website here) .com you don't know what you're talking about.

need i say more?


i'm sorry guys it's nothing personal, but do feel compelled to defend what you prefer? don't you ever get tired?

and kex put it best when he said:

grrr brain cant take this again

we need sticky
another good quote:

Xeon or Opteron...

#1 suggestion:

Search Button... this is a weekly topic.
again, it's nothing personal, but just an observation about these threads in general. :shrug:

[JasonA]

take a trip with me and see what the future of this thread will look like...

..etc
..etc

need i say more?
LOL

so true it seems
:scream:

[Dougs]

You should tell him to buy both and let him figure it out and then he can tell us which one he thinks is better.

[elvis]

I wouldn't go with an opteron system. According to Tom's Hardware Guide the Xeon out performed opteron in every 3d benchmark test. If you don't believe me read the article—"Battle of the Titans Xeaon VS Opteron" at www.tomshardware.com (http://www.tomshardware.com) .
Tomshardware is cash for comments. Go read a real website like xbitlabs, anandtech, arstechnica or something equally as reputable.

People who buy hardware based on Tom's deserve all the problems they get.

I had the pleasure of testing a dual Opteron 244 (1.8GHz) setup on an Iwill DK8N and a dual Xeon 2.8GHz setup on an Asus P4CDL just this morning. At 1GHz slower, the Opteron setup managed to match or beat the Xeon setup. I know where I'll be spending my money on my next upgrade.

[Fahad]

i should run a psychic hotline

[elvis]

Sorry folks. The hairs on the back of my neck all stand on end when someone starts quoting tomshardware as the one and only reference on the web.

I wouldn't buy a piece of kit until I read a dozen reviews AND played with the physical hardware myself first. I still don't understand how people can read one review and then spend their hard-earned cash based on that. You may as well go bet it all on the horses for the same result.

Read a review. Then read 10 others. If they all concur, then go play with the hardware on some demo gear before making up your mind. Don't take the advice of a single website (or single forum post for that matter) as "the word of God."

[alanmac]

Whilst I agree with Elvis on not taking one review as gospel and find out as much as you can from different sources.

But the possibilities of being able to demo a particular piece of hardware let alone running the applications you will be using to make a comparison are remote to say the least.

Practically speaking the best we can hope for is either a collection of reviews and maybe first hand experience from other users, even then can we guarantee that the user has their system running at optimum performance?

I think you'll agree that every link in the chain is important and the whole system must be balanced and working well together. No use buying the best CPU you can get your hands on if you saddle it to a ill matched hard disc or graphics card within your system.

Matching and balancing your components I would say is far more important than just choosing one type of processor or the fastest and biggest of everything.

The reason that most specialist computer manufacturers charge more than mainstream suppliers such as Dell, HP, etc is the time and effort they put into putting together a system that works well and to its best performance for a particular software use.

That does not mean to say you cannot put together a brilliant system yourself if you have the time to research the components, troubleshoot any problems, configuration issues and feel confident in all these aspects.

I would be most surprised given the cost of individual components that building the equivalent system yourself from new parts to what's available "off the shelf" would save you any money given the purchasing power of most big manufacturers. Notice I say equivalent.

Whatever you choose it will be very fast - but "out of date" by this time next year, if your lucky, such is life in the fast moving world of computer technology.

That does not mean to say you can't keep using it, enjoy what you do with it and create great work on it !!

Just my opinion

good luck with your choices

Alan

[deumilcat]

take a trip with me and see what the future of this thread NOW will look like...


no, according to (enter username here) "Tom s Hardware" is (enter arguments, quote, -ir-relevant experiences), so please (some condesending statement about the other guys knowledge and sources)

no, according to (enter other username here) (enter uber tech website nobody knows about yet) is (enter arguments, quote, -ir-relevant experiences), so please (some condesending statement about the other guys knowledge and sources)




lol this thread turned into "Tom s Hardware" or "other underground hardware websites":applause:

[lots]

*smirk*

;) hehehe..

[Fahad]

take a trip with me and see what the future of this thread NOW will look like...


no, according to (enter username here) "Tom s Hardware" is (enter arguments, quote, -ir-relevant experiences), so please (some condesending statement about the other guys knowledge and sources)

no, according to (enter other username here) (enter uber tech website nobody knows about yet) is (enter arguments, quote, -ir-relevant experiences), so please (some condesending statement about the other guys knowledge and sources)




lol this thread turned into "Tom s Hardware" or "other underground hardware websites":applause:

and here i thought i was the only one who can see the future :thumbsup:

btw.. i think there should be one thread that can be soley used for wars about anything at all, and only one thread.

[elvis]

take a trip with me and see what the future of this thread NOW will look like...

no, according to (enter username here) "Tom s Hardware" is (enter arguments, quote, -ir-relevant experiences), so please (some condesending statement about the other guys knowledge and sources)

no, according to (enter other username here) (enter uber tech website nobody knows about yet) is (enter arguments, quote, -ir-relevant experiences), so please (some condesending statement about the other guys knowledge and sources)

lol this thread turned into "Tom s Hardware" or "other underground hardware websites":applause:

Read between the lines, friend. The underlying message is don't take ONE SINGLE source of information as gospel. That also goes for anything I have to say. If someone buys hardware based on my advice alone, it's just as bad.

Yes, personally speaking I despise tomshardware. What I despise more is people who continuously only ever take one website as the be all and end all. Everyone makes mistakes. Anandtech is what I consider a reputable source of information, and only 2 weeks ago they pulled an article due to discrepencies in their testing. The took all of their test cases back to the lab and re-tested from scratch, releasing version 2.0 of the article complete with new and updated data.

Now had the end user bought their equipment based on that one article, they could have possibly wasted their cash. Had they read around, they would have probably noticed differences between the mean results and Anand's.

Again, not everyone is perfect. One single source of information is a risky way to base your purchasing decisions. Toms, or otherwise. You wouldn't buy the first house a real-estate agent shows you. You wouldn't buy the first car you test drive. You wouldn't bet all your money on one horse because old uncle Bob told you to. So why buy your hardware because of a single website? Sheer stupidity.

Get out there and read. Learn. Improve your own knowledge. It's free, and more to the point it may end up saving you money.

</soapbox>

[Horik]

If you would actually read what I wrote, you would see- "I WOULDN'T BUY AN OPTERON SYSTEM". I was making a personal statement. I didn't tell anybody to go out and buy one thing or an other. Tom's Hardware Guide is a good source, and I would like the guy who stated that it is a cash site to prove me wrong. Show me 10 independent sites that say the Opteron is a faster processor then the Xeon. You could argue that they all are cash sites. These forums are for our personal opinions and thoughts, so quit being a dick.

[maninflash]

I did these render tests a while ago using the scene from zoorender at:
http://www.zoorender.com/benchmark/Benchmark_Mental.zip (http://www.zoorender.com/benchmark/Benchmark_Mental.zip)

Both operating systems are Windows XP pro with Maya 6.0

00:41:75
Dual XEON 3.2GHz, SuperMicro X5DA8 mainboard, 2GB RAM, with Hyper-Thread

00:47:68
Dual Opteron 248, Tyan Thunder mainboard, 2GB RAM

00:50:88
Dual Xeon 3.2GHz, SuperMicro X5DA8 mainboard, 2GB RAM, without Hyper-Thread


with HT on, mental ray renders about 9 seconds faster in a single frame. That's because, mental ray is a multi-threaded renderer and it can use as many cpus as you have, you should enable this in the mental ray options by choosing "using 4 threads", which on a dual xeon, is four, two physical and two virtual created by HT, but mental ray doesn't know that and so thinks you have a quad system!

On the other hand, Opteron has better memory bandwidth, which helps a lot when your scene gets heavy or when compositing higher resolutions images. If you use a linux distro that supports 64bit, you can have more than 4GB of memory on an opteron box which would help on certain renderers like mentalray standalone version or pixar renderman.

[lots]

You also have to keep in mind that certain types of renders will perform differently on the two different architectures

something with alot of volumetrics for exmple, or hair, or what have you can change the out come of your tests.

Not taking any sides here cuz this is just a silly thread. I'm just pointing out that to test properly you need a wide range of data to really back any of your claims up. Something that not many of the big benchmarking sites really do. At least when it comes to testing speed of rendering within a certain renderer on each CPU platform ;)

[CGTalk Moderation]

This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.