- Thread Starter
- #21
AMD were running in supercomputers not because their performance per core was higher or total computation pperformance per CPU was higher but because it was possible to put more cores ( CPU sockets ) on a single board. Simply more cores on the same space were giving that advantage in calculations. Not to mention much lower cost per core. Other thing is that earlier AMD had lower delay on data transfer between cores so they're actually good for many small operations at the same time.
Before Core 2 generation, Intel was better choice as it was more reliable. After it was better choice as it was offering higher performance in most operations. Saying about reliability I mean quality of motherboards, availability of spare parts etc. which was always lower for AMD.
Since I'm working in IT ( 12 years or something ) I have never seen that AMD had more than ~15% market shares on EU market in business products. AMD was always related to home entertainment and people actually see it like that all the time.
Right now I don't think anyone will choose AMD over Intel for work. AMD business department is like dead. The same in servers and laptops.
Personally I don't know single distributor in central EU which is offering AMD based servers from stock and I work in distribution for couple of years. In Poland I haven't seen AMD based server for like 6-7 years. Simply people don't trust AMD to buy them for business, not to mention lower performance etc. comparing to Intel.
The biggest server manufacturers don't have anything based on AMD in mass production. Couple of years ago IBM was keeping 1 ... literally 1 AMD server series because of some single operations which were running faster on AMD and that was when first Phenoms were on the market.
Yes very true, the core count was what drove AMD server side for a long time. Core count can help in a lot of ways to process multi-core programs that are integer based.
I run cfd programs, why can't any of my software use the hyper threads on my intel cpu's but can use all 8 cores of fx?
how the hell do they fit over a billion tranzisters on a chip?? that's a billion somethings, I can hold in my hand!!!!
folding at home uses our gpu's to compute stuff, I just bought some very high dollar software that will be running on quadro 6000m's, will this type of software we comeing to the home pc or is the cost just to much?
holly mother of god, I could fly with creflo dollar for the cost of this stuff over the next ten years!!!!
So many questions!
CFD Programs: These programs rely on floating point operations to solve huge differential equations in several different integrals. A couple of my mech-e friends took fluids, their stories gave me night terrors. Whether your AMD or Intel, you have 1 FP unit per a core. Hyper-threaded CPUs does not mean there is another set of hardware to run instructions on. Rather, its a clever way of sneaking instructions between instructions already setup in a que. Think of it like Disney World's Fast Pass system. Someone in a completely different part of the park is waiting in line with you, but you don't know it. Right before you get on the ride, they jump in front of you and enjoy the ride with a brighter smile on their face than you. Pretty much the same is happening. The Intel's CPUs think very hard on how they place instructions in the que, and when an instruction should be accelerated.
Back to the CFD problem, the reason why Hyper-threading does not work is because the CPU cannot find a faster way to already process the instructions its fetching. Also, if an instruction did use HT, it would end up using the same core anyway. You are most likely creating overhead on your system when you have Hyper-threading on; at least this was true back in P4 to Nehalem days.
Why does AMD 8core get utilized but not Intel? I already went over Intel, AMD is a bit different. Since their cores are actually two logical processors they can execute more instructions. AMD's Bulldozer architecture is unique. Instead of cores AMD calls them Modules. Two cores share a front end to assign instructions to the two integer cores. These two cores also share a FP core. Thus a Module has one front end, two integer cores, and one FP core. Your CFD calculations will also require simple instructions to move data around the system. Integer cores will handle this, while the FP core is executing instructions. This does not mean that AMD CPUs are more efficient than Intel. Although AMD theoretically can execute more instructions, the quality of performance is less than that of Intel's cores.
More on AMD vs Intel in previous posts:
AMD Modules vs Intel Hypter-threading
Transistor Count: I'm going to answer this in another post.
CFD Software: I'm not familiar with these types of programs. I kinda made sure I never got near them because the math scares me. Math based programming is not my strong suit. Now speaking with experience with industry standard tools, most of them will never get to the point for consumer grade computers to be ran on efficiently.
Those Quadro's will do some work with your programs. GPUs compute FP efficiently due to their large number of cores.
That software costs a lot because very smart mathematicians and programmers figure out the best algorithms to efficiently display those results.