http://www.hexus.net/content/item.php?item=4843&page=3 Granted this is using mega expensive hardware and there may be other things at play here that make for the difference but I must say Intel has really gotten their act together if those results mean anything. Especially the fact that a slower clocked Intel was actually able to outperform a OC'ed FX 60. Crazy stuff.
I knew that Intel is very rich company and will leave behind the AMD soon. Just look here they are out with quad core processor now http://www.informationweek.com/news/showArticle.jhtml?articleID=181502198
About time, I guess they felt they should actually start trying after AMD is sucking up all the market as of lately. Better late then never I guess, let the wars begin again!
Let these companies fight......We will get the benefit. Always old products value goes down when new one comes in the market and if these companies fight like this then new and speedy processor's will come out more often.
If this trend continues, then we'll see the 2000-2005 era all over again. Intel win, AMD lose, intel lose, AMD win. I'll also be interested in what Anand makes of this.
I reckon it'll be like that but with cores! who can get the most cores into one CPU - although this could require frequent socket upgrades to make the most of more cores.
Cores? I doubt it, its easier to increase clock speed than whack another core onto the die. It would have more effect on heat output than clock speed. Maybe internal CPU communications will do.
Well I would expect Intel to prep for the Pentium 5 soon enough, I mean the P4 has been around for ages and has seen a 3GHz increase from it's start, my guess is around 2k7 they will have a substantial enough architecture revision that they can dub it as something new. As far as AMD goes it's even harder to tell as last time I checked there wasn't anything major ahead on the roadmap.
Well all I know is they won't take this conroe business lying down. Sure humiliating to have a $800+ big daddy proc loose hands down to a $530 Intel who's always pushed the importance of clockspeed but I can bet they've got something up their sleeve.
After reading a few opinions, I should say this. The boxes supplied were by Intel, they had prebuilt the custom Conroe system with an engineering sample and an optimized engineering motherboard. Crossfire was used, not SLI which is better suited to AMDs platform. "As far as we could see the systems were fairly setup, but please note these tests were not conducted in the HEXUS labs and are a mere indication of what performance we will see. " from Anandtech " (the ATI graphics driver was modified to recognize the Conroe CPU but that driver was loaded on both AMD and Intel systems)." A graphics card that needs a special driver to recognize a CPU? Didn't know that. While I'm not going to say that those benchmarks were wrong, there is no proof of that. Seeing the way those pcs were set up, don't know abot you but it rings alarm bells with me. I'll wait until i see production chips being compared independently before I will accept that there is a so called 20% performance increase. This is not fanboyism (despite my avatar), this is skepticism.
AMD had announced quad-core before Intel, IIRC. I know Intel had trouble getting a 4GHz chip to remain stable within certain parameters so they decided to drop that. I have no doubt that if Intel had been able to get a 4GHz or faster CPU within the right parameters, there'd be one out now. Don't expect to see a Pentium 5. From what I'm understanding, Intel is dropping the Pentium, and eventually, Celeron brands and just going with CPU models 3xx, 5xx, 6xx, 7xx, 9xx. The first number would be the market grade (the 300's would be the Celeron, everything else would be a Pentium-class). The higher the first number, the more features it would have. The higher the last two numbers, the faster the CPU's clockspeed would be. Time will tell how this bowls over. Everyone thought AMD's PR rating system would be the end of it all, but they've run pretty well with it. Once websites can do their own internal testing, we'll see how Conroe performs. However, I believe they're taking lessons from their excellent performing mobile chip, Dothan, and moving that to the desktop. The architecture is more effiecient and effective than anything under the Pentium 4 probably has been. Unfortunately, nVidia hasn't officially released a universal SLI driver to work on non-nForce chipsets, and I think if they don't, that could cost them dearly. ATi's Crossfire could use some work, but, unlike SLI, you can run a Crossfire setup on a non ATi-chipset system.
