Dear forum members,
We are delighted to inform you that our forums are back online. Every single topic and post are now on their places and everything works just as before, only better. Welcome back!
Discussion on Article:
Six Cores from AMD: AMD Phenom II X6 1090T Black Edition and Phenom II X6 1055T CPU Review
They are no hardcore folders out there that use AMD processors. I have 2 AMD processors just to support my GPUs and that's it. It's simply not worth the money to fold with AMD CPUs.
xbitlabs: did you even try to overclock the 1055T? you end up kind of recommending it for overclockers based on your results with a different processor which is not multiplier-locked...
End of story.
You have decided that AMD's 6 cores are worse than intels 4 cores based on useless synthetics like 3dmark vantage? Or worse, games which use two cores? That doesn't add up does it?
Back to review school for you - start by reading the other reviews on the internet to see where you went wrong.
The rewiew sites owe it to us and AMD to review the chip fairly, not benchmark it with a bunch of poorly threaded games and synthetics then declare "Intel quads are faster clock for clock".
Of course the intel quads are going to be faster clock if clock if you benchmark next to nothing that uses the AMD chips full 6 cores.
You guys make one of the best reviews in regards to many things; PSUs, Power Consumption, Motherboards, Coolers and of course LCDs, best in the business. On the CPU testing front you should improve your methodology, as your usage scenario hardly suits the way your readers use their CPUs (and maybe you too for that matter).
I'll illustrate with an example what I mean: say at 100$ mark we compare a C2D e7500 and an Athlon II 630; your testing results will say in a nutshell: Athlon II is for highly multithreaded apps, while the e7500 offers higher speed for not so multicore optimized apps. You compare the CPUs per app, but we don't use our CPUs like that any more. Should you address this aspect? I think so.
There is one usage scenario that isn't covered: users that use many apps - and with multi core CPUs knocking on our doors this usage scenario is increasingly more important. Add a second monitor, and in parallel to Crysis you can browse heavy-flash web page(s), while downloading stuff and keeping track of your friends thru facebook and twitter apps. With the antivirus security set on "high" would you venture to say Crysis will run better on the dual core with fast cores or the quad with more albeit slower cores? Most users "feel" their quad is faster than their former dual core, even in older games, although reviews say otherwise. There must be a way one can quantify this "feeling", this added snappiness, and you are the ones for this job in my opinion.
To finish my post I'll place a link of Anandtech's server benchmark - in house build to assess virtualisation performance in "few but heavy" virtual machines installed. They used a database as departure point for bench, you may as well make a pool and gather data on who uses what simultaneously, make a batch and then make a score starting from the per app data gathered: FPS, seconds to completion of AV/encoder, etc. Using SSDs and a home-based web/dlding server should make scores repeatable.
You mean Core i5 750. There is no such thing as Core i7 750. Also you make assumptions that just because i5 750 doesn't have hyperthreading that it'll be far worse than i7 930 which is about matching AMD's X6 or doing better. Maybe you should have tested it first. Also let's not forget that if you disable HT, you should be able to overclock even higher, negating any such possible reduction in speed.
Also lol at the other comments. x264 is a highly thread optimized benchmark. That's what you should be looking at, not the useless synthetic benchmarks or the games that use 1-2 cores.
Also, there is an image on page 2 that shows 2 72-bit DDR2 memory channels coming from the IMC. I think that this is an Istanbul Opteron slide that uses ECC RAM. Not that big-a-deal but I thought it was out of place.
Seems the author had already decided back then that Thuban wasn't going to beat the Nehalems.
Needless to say, this is *incredibly* poor form xbitlabs. Get another - preferably unbiased - reviewer and review this chip properly because this one is a farce.
i remember AMD started woopin Intel's butt when they introduced the IMC(integrated memory controller) and got rid of the FSB, no middle man between the cpu and ram. crappy proc w IMC will beat good proc w a slow middle man.
Kinda funny in 06 core 2 line (Conroe) comes out and beats the hell outta everything AMD had at the time. Guess what ....the Core 2 Duo/Quad line up all still used a FSB middle man w no IMC. Hillarious......
Now since AMD has finally beaten the older Core 2 lineup (lmmfao)they need 6 cores to go against 4... is it fair? 6 vs 4? lolz comparisons should be 6 vs 6 or the quads vs quads... marketing mumbo jumbo u gotta filter thru it all to get the little bit of logic.
Remember now, AMD has had to lisense ALL cpu instruction sets from Intel or BUST! ... i.e. x86 mainly w all the SSE sets. AMD 64 is an iteration from IA-64. (Itanium procs by HP and Intel)
AMD wouldnt be alive if it wasnt for the court ruling in 1991. Allowing them to basically"steal" Intel's technologies and "clone" their procs for cheaper. Making the Am386. IBM had a big hand in this issue as well. Must be great not being original at all....= AMD
But im for cheap as well but performance...is the main issue, along w "future proofing"
Fanboy of neither btw....... originality pwns all.
the i7's are basically revamped core2's w QPI and hyperthreading and some instruction set optimizations. So it goes both ways....
yes they do look promising i agree. but they are still playing catch up ........ i wonder what type of tech we would be playing with IF we had other "original" companies to level the playing field and force each other to keep coming up with newer better things at a faster pace. Mobile and Desktop wise.
Sure keeping same mobo, psu, and everything else system wise for some random amount of yrs to save money sounds awesome....... nah it doesnt rly........ things could be so much diff right now for us if we werent with this mindset the markets have created. Slowing technology down, and hindering our technological evolution as we advance, by using this "budget" scheme of things. So none of this really matters..... Be stuck doing same thing for yrs on end because the companies only care about "budget", not advancing. NVM that subject lol
AMD needs to be "original" again, since theres only 2 "major" cpu manufacturers, so that we can advance into newer better things. The more as tech advances, the more we do, but only allowed as far as the market greed will let you. (like Intel always being $500 and up for "good" So dont take my posting wrong. LOL
I hope I got that out there right. lol
Add your Comment
Enter your username and e-mail address. Password will be sent to you.