barton core
- Thread starter Pablo11
- Start date
-
- Tags
- CPUs Power Supplies
rain_king_uk
Distinguished
- Nov 5, 2002
- 229
- 0
- 18,680
Because AMD felt that the extra cache justified tinkering with their CPU rating.
Big mistake IMO - unless there are technical reasons not to they should have clocked more and been less optimistic with their new scale. Because of the lower clockspeed and optimistic rating Barton looks like a bid disappointment in the benchmarks, when actually it's a nice step forward.
I'm guessing perhaps the reason they didn't clock it higher is because they didn't want to risk Intel releasing a new faster chip to compete with it.
<P ID="edit"><FONT SIZE=-1><EM>Edited by rain_king_uk on 02/21/03 07:05 PM.</EM></FONT></P>
Big mistake IMO - unless there are technical reasons not to they should have clocked more and been less optimistic with their new scale. Because of the lower clockspeed and optimistic rating Barton looks like a bid disappointment in the benchmarks, when actually it's a nice step forward.
I'm guessing perhaps the reason they didn't clock it higher is because they didn't want to risk Intel releasing a new faster chip to compete with it.
<P ID="edit"><FONT SIZE=-1><EM>Edited by rain_king_uk on 02/21/03 07:05 PM.</EM></FONT></P>
I suspect they are having a big problem with manufacturing chips at 2.25GHz. This would explain why the Thoroughbred 2800+ has been virtually non-exhistant since it's paper-launch several months ago. So in order to avoid the same criticism, they decided to launch the Bartons at lower clockspeeds and refer to a few carefully selected benchmarks where it peforms better than the P4 3.06GHz. However, as most of us know, this plan has backfired because the Thoroughbred 2800+ outperforms the Barton 3000+ in many benchmarks and the P4 3.06GHz still blows it away in most benchmarks.
I agree completely this was a big mistake for them. Their PR numbers are quickly loosing credit and are being seen as a marketing scam to fool the public.
Listen to me now and forget me tomorrow
I agree completely this was a big mistake for them. Their PR numbers are quickly loosing credit and are being seen as a marketing scam to fool the public.
Listen to me now and forget me tomorrow
Speculation time:
Downdside:
The Barton Core seems to be flat out slower than the ThouroughBred core, though it's likely just the lower clock speed. The performance ratings are blown out of proportion.
Upside:
The Barton Core seems to have a faster memory interface. Synthetic benchmarks point in this direction.
The P4 accels in applications that are adjusted for higher clock speed (actual MHz rating). These applications dominate the market.
Seems to me, these same applications are the ones putting the ThouroughBred ahead of the Barton. Many benchmarks where AMD is ahead of Intel also has the Barton ahead of the ThouroughBred. However, that isn't many benchmarks.
Only 3 (I think) in the latest roundup of 65 processors at Tomshardware.
I'm guessing Barton can be overclocked more than the ThouroughBred, because it has a faster memory interface. I've read a couple of sites that comment on its excellent overclockability, but none really speculate on why.
I wonder if someone could get a Barton up to 400 - 433 MHz FSB. Paired with the same speed memory, it may finally justify PC3200 and PC3500 memory.
I am in no way justifying the XP ratings of the new Bartons. They may well be overly optimistic, or they may (likely) be a scam. Either way, you'll not catch me buying AMD anytime soon.
Before I get too harsh on AMD, I've seen similar practice from Intel. AMD wouldn't be in the situation they are in right now it intel hadn't screwed them with the P4. Intel bumped the MHz ratings going from P3 to P4, but they dropped performance. They proceeded to market the P4 as the equivalent or better to the Athlon and P3. A P3 1.4 GHz is easily faster than the P4 1.6 and provides healthy competition with the P4 1.8 GHz. That's a 28% increase in MHz with debate over any increase in performance.
AMD's XP3000+ is usually faster than the XP2700+, but we'll call them even for the sake of my point. This is only an 11% increase in rating with debatabe over any increase in performance.
In any case, I'll watching the Hammer with great interest and criticism when it is released. AMD still isn't bumping up clock speed like they should. With most apps now based on clock speed and SSE2, the Hammer will struggle much more than if it were already released.
Pain is the realization of your own weakness.
Downdside:
The Barton Core seems to be flat out slower than the ThouroughBred core, though it's likely just the lower clock speed. The performance ratings are blown out of proportion.
Upside:
The Barton Core seems to have a faster memory interface. Synthetic benchmarks point in this direction.
The P4 accels in applications that are adjusted for higher clock speed (actual MHz rating). These applications dominate the market.
Seems to me, these same applications are the ones putting the ThouroughBred ahead of the Barton. Many benchmarks where AMD is ahead of Intel also has the Barton ahead of the ThouroughBred. However, that isn't many benchmarks.
Only 3 (I think) in the latest roundup of 65 processors at Tomshardware.
I'm guessing Barton can be overclocked more than the ThouroughBred, because it has a faster memory interface. I've read a couple of sites that comment on its excellent overclockability, but none really speculate on why.
I wonder if someone could get a Barton up to 400 - 433 MHz FSB. Paired with the same speed memory, it may finally justify PC3200 and PC3500 memory.
I am in no way justifying the XP ratings of the new Bartons. They may well be overly optimistic, or they may (likely) be a scam. Either way, you'll not catch me buying AMD anytime soon.
Before I get too harsh on AMD, I've seen similar practice from Intel. AMD wouldn't be in the situation they are in right now it intel hadn't screwed them with the P4. Intel bumped the MHz ratings going from P3 to P4, but they dropped performance. They proceeded to market the P4 as the equivalent or better to the Athlon and P3. A P3 1.4 GHz is easily faster than the P4 1.6 and provides healthy competition with the P4 1.8 GHz. That's a 28% increase in MHz with debate over any increase in performance.
AMD's XP3000+ is usually faster than the XP2700+, but we'll call them even for the sake of my point. This is only an 11% increase in rating with debatabe over any increase in performance.
In any case, I'll watching the Hammer with great interest and criticism when it is released. AMD still isn't bumping up clock speed like they should. With most apps now based on clock speed and SSE2, the Hammer will struggle much more than if it were already released.
Pain is the realization of your own weakness.
Hammer struggling with SSE2:
My understanding is that Hammer will have SSE2 instructions. Therefore, applications optimized for SSE2 shouldn't hurt Hammer too much. I agree about the mHz, though. When Intel first went to the .13 process, it was a while before they were really able to ramp up the mHz, and they have far more fabs than AMD. I think AMD can probably ramp their process up to around 2.5 or 2.6 mHz. We'll see if that's enough for them to stay with Intel. A nice bump to a 400mz FSB wouldn't hurt either.
I want to move to space, so I can overclock processors cooled to absolute zero.
My understanding is that Hammer will have SSE2 instructions. Therefore, applications optimized for SSE2 shouldn't hurt Hammer too much. I agree about the mHz, though. When Intel first went to the .13 process, it was a while before they were really able to ramp up the mHz, and they have far more fabs than AMD. I think AMD can probably ramp their process up to around 2.5 or 2.6 mHz. We'll see if that's enough for them to stay with Intel. A nice bump to a 400mz FSB wouldn't hurt either.
I want to move to space, so I can overclock processors cooled to absolute zero.
TRENDING THREADS
-
Question User RX 580 8gb red devil powercolor, asking original BIOS- Started by PizzaInGame
- Replies: 5
-
-
-
News Microsoft updates Windows 11 24H2 requirements, CPU must support SSE4.2 or the OS will not boot- Started by Admin
- Replies: 19
Facebook
Twitter
Instagram