I thought I would ask which will be more powerful in terms on gigaflops. I know that with AMD motherboards it is possible to have quad sockets where as with intel montherboards you can only have dual sockets on your motherboard. So if you were to compare the latest processors on the date 1/12/2012 then which would be more powerful and more efficient when comparing the AMD quad socket to the intel dual socket. Also what sort of specs are speculated for the biggest processors due by 1/12/2012.

In case your wondering why I ask it is because I am going to be buying one of these babies and I want to be able to render high definition images on progressive path tracing render. In addition I like to process large algorithms which take up enormous cpu like a project I have done recently with my i7 cpu which was when I downloaded the wikipedia project and a few of the other wikimedia projects along with the guttenberg project and put them into a database for processing so that a user can place into a textbox a question and the computer will be able to answer the question in a 2048 character or less string by filtering the database by relivence. Obviously I didn't have the cpu but am waiting for these new cpu releases the come out so I can use such algorithms that I produce. So please descuss on regards to the original question. Thanks. :)

Dani AI

Generated

A few clarifications up front (date and platform): if “1/12/2012” is January 12, 2012, AMD’s Bulldozer-based FX desktop chips had already started shipping in late 2011, while Intel’s consumer Ivy Bridge desktop CPUs didn’t launch until spring 2012. That timeline matters because the quad-/multi-socket options you hear about are generally server-class parts rather than the desktop FX/Core families. (en.wikipedia.org)

What the microarchitecture means for floating-point work: Bulldozer uses a modular (CMT) design where a module exposes two integer schedulers but a shared “FlexFPU” (paired 128-bit FMAC units). That helps some highly threaded integer workloads but can reduce sustained FP throughput when both threads on a module are FP-heavy. Ivy Bridge-style cores are full per-core designs with dedicated FP resources and generally higher single-thread IPC — which often wins in workloads that are not perfectly parallelized. As and pointed toward, you’ll see this behavior show up in real-world benchmarks. (en.wikipedia.org)

Why raw gigaflops isn’t the whole story: usable rendering speed depends on FP units, SIMD/AVX support, memory bandwidth, cache, and how the renderer parallelizes work. For progressive path tracers, GPU renderers have long offered far more usable GFLOPS per dollar and better interactive/progressive previews; many production and interactive engines are GPU-first and scale well with multiple cards, so a GPU-focused workstation or a small GPU render-farm is usually the better investment for HD path-traced rendering. (cgchannel.com)

Practical recommendation for your stated goals: if you need fast progressive path tracing, prioritize one or more modern GPUs with large VRAM and good driver/support for your renderer, and pair them with a CPU that has solid single‑thread performance and plenty of system RAM (ECC if the dataset is critical). If the priority is CPU-side large-scale text processing and indexing, prioritize many cores plus lots of RAM or use a distributed/cluster approach — multi‑socket boards are server hardware (Opteron/Xeon class) and not usually the most cost‑effective route for a desktop render/compute workstation. (en.wikipedia.org)

Recommended Answers

All 3 Replies

Its almost on par with the 2600k, and just a touch cheaper than the 2600k, I waited for months for them to come out as I had plan for a rendering farm using them, on paper, 8 x 3.6GHz sound like a dream for the price, but then, the term 'you get what you pay for' came out so true, sticking with our Q9650s until Ivy.

Maybe you should look at a few comparisons.

I was hoping to buy one of the Bulldozer cpu's, but looking at some of the results, I will be sticking with my 1090T 6 core.

http://www.anandtech.com/show/4955/the-bulldozer-review-amd-fx8150-tested/7

The link provided tests the AMD 990FX however doesn't the AMD Bulldozer include AMD's version of hyper-threading along with better caching, lower power consumption and higher clock speeds? Thanks.

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.