Originally posted by: IntelUser2000
I say that 15% is a huge disappointment simply because it took an enormous team of engineers several years to reach that goal. I'll bet they would have made more gains by having those people research manufacturing and chipset design.
If K10 ends up 20% faster than K8 . I think the engineers did a great Job. On the low clocked K10 power consumpsion looked good. K10 is looking good. I think to many people had to high of expectations for this old core. Really how far back does this core really go? Links please!
Though you can say part of A64's success is Intel's fault with Prescott, and part of the reason Barcelona looks bad is because Intel exceeded everyone's expectations with Conroe. Nobody expected Conroe's performance before the numbers started trickling out. Everyone expected that with same FSB, it would be just like a Yonah. However, we have seen its much more than that. People also doubted Intel's ability to clock Conroe cores to more than 2.5GHz, thinking of Yonah. They came out with 2.93GHz EE.
Even tho I am new here . I have read the forums an articles here for years. I do recall a woman that actually called the Merom performance numbers right on the head. I tried to find the threads but can't remember her name. But she actually after the facts came out . Pawned all.
You can say this is like a Geforce 5800 vs Radeon 9700 comparison. Geforce 5800 would have been impressive if ATI didn't manage to pull out an even more impressive R9700.
Very true
Someone may have already said this, but on the desktop side, few people really NEED the power of a core 2 duo cpu. I would say gamers and other enthusiasts are the only ones that actually take advantage of a chip like that, and they make up what, like 10% of the market? I just bought my girlfriend a laptop with a 1.9 ghz turion x2 and 1 gig of ram, and it plows right through vista, multitasking, movies, etc. Which is what most of the market uses a pc for.
If you have the force one must learn how to control it
Yes, of course, so why are you posting here at all?? It's about people that actually cares about the difference that's posting here. After all, if people buy what they only need, Intel wouldn't have been doing suddenly better by introducing Core 2, and AMD wouldn't have been in trouble.
Exactly
It's not unlikely to think that Phenom scores will improve by at least 10%. Memory gains from using higher speed non server DDR2 667 RAM should be somewhere around 5%. Mature chipset drivers catered specifically for desktop apps should yield another 5% as well. These preliminary scores are very promising in the server arena, but if AMD wants to regain lost share in the desktop arena they better pray they can get those clockspeeds up.
I would say between 5-10% is realistic
That sir, won't happen. Some people seem to have forgot Anand's early Conroe review at IDF, and how people complained of BIOS not recognizing K8 CPU and its a crappy chipset etc etc. We found out later how much the latest BIOS brought to the table. An average of 1-2%.
Well one could always go and get those early reviews and compare them to the up and coming X38 reviews . I think you will find that 2-3% is way low est.
About chipset and memory:
http://www.anandtech.com/cpuch...howdoc.aspx?i=3042&p=6
Look above, look how irrevalent chipsets are to performance. Unless we are comparing totally ignored chipsets like Via's and SiS's, chipset, performance difference between the slowest and the fastest chipset don't exceed 3%, and majority of the time its under 2.5%. And one of major changes to Barcelona/Phenom is greatly enhanced memory controllers. If anything, it will benefit less than Opteron does from faster memory.
People like to dream .