Will my PSU handle Crossfire 4830's?

Status
Not open for further replies.
well the problem is that K10 is really just a slightly modfied form of K8 which is a great architecture but it has been around for too long now.

I wouldn't go as far as saying as theres no hope for amd, because the chips are not slow...theyre just not as fast as intels.

Now deneb does look very good from all the stuff ive seen. It doesnt have to beat nehalem for me to move to it. I just want good performance from an amd quad. not looking to break world records or super pi here.

Apparently theres some chip called "bulldozer" in the works which is a whole new arcitecture (and cpu socket i would imagine) so 2009 wil be an interesting year.

[H] Enthusiast - AMD's Direction and Next-Gen Bulldozer Revealed
 
I am more hopeful about Bulldozer. Wiki claims its all new from the ground up.


But look, I'm not being a fanboy here. Intel tried die shrinks, cache increases, higher clock speeds, everything to boost P4. it never worked because ti was a slow architecture. Phenom has the same problem, i don't care how much cache you add, or how high you clock it, its still going to be slow. that much is evident. if Intel couldn't throw money at netburst and make it a champ then what makes you think AMD can?

You may not need a nehalem beating chip from AMD to buy it, but most people do, and AMD does desperately if they want to make money. Deneb wont be cheap, so compared to nehalem and even current Core2 chips the price/performance is shot.

K8/K10 has met the end of its life, lets be honest with our selves. AMD gambled on Intel trying to release Tejas but they went to Core 2. AMD should not have rested on their laurels and should have kept going ahead with newer designs. Phenom is a flop. It is cold bugged and its slower than Core2 across the board. Corei7 is twice as fast as the best Agena, and Deneb has been shown to marginally faster clock for clock than Agena. AMD should know by now, after what Intel did, that you cannot overclock your way out of a hole.
 
Same applied for Intel during Pentium 3 and Athlon XP, by the time they had P4 out, they had higher end versions of the XP series.. leading us here to this day with Core 2 Duo...

I don't know if the same applies but there's still slight hope for them, just like there was for Intel

One of the main reasons Intel was able to recover from netburst with Core 2 is they had parallel design teams working on Pentium 4 and Pentium M. So when they realized netbusrt was never going to beat K8 they had a second architecture to turn to.

Unfortunately AMD doesn't have that advantage so it will be more difficult for them to catch up with Intel.
 
And they don't have as much cash to put into R&D. Even though Intel's market share took a hit, the still turned a healthy profit. I wish the same could be said for AMD.
 
So what does the future hold...... AMD always being a step behind Intel?

But that makes no sense, because then it will be the end of AMD.

Intel will get to charge as much as they like for their CPU's then.............. no more price/performance.

No I m not bashing Intel, its just the realization that Intel will start sitting on their asses with nothing new, and innovative. Until another company gives them a run for their money like AMD had.
 
If AMD wishes to survive they will need to pull a winner out. They have done so in the past. Historically they were always able somehow to make faster processors for less. This is the first time in a long time that they have not (probably why there are so many AMD fanbois) Bulldozer may be a winner, but there is no way to know now. In my opinion Phenom is a bad architecture if you are trying to sell chips competitively and it should be dropped at once.
 
There was no hope for the p4 to best Athlon 64, as there is no hope for phenom to best Core2 or Corei7. The change was caused by a new architecture. AMD must abandon K10 and make something new. K10 is inherently flawed and trying to fix it wont help, you have to start from scratch.
The only flaw was trying to rush it out before it was really ready.
But, AMD had to do something. And so they released it. In 65nm, with the TLB bug, and in not a particularly well optimised form.
But, Deneb has more than its fair share of enhancements to break it away from Agena.
That includes much better performance per watt, and overclockability, as well as a good IPC increase. It's the same architecture, with some enhancements.

I'd say that it will at least compete well against Yorkfield.
 
amd needs to be putting all their effort into a whole new socket and core design like the bulldozer and be sure that its not released prematurely with bugs or bad performance.
 
Status
Not open for further replies.
Back
Top Bottom