Everyone keeps saying the Palomino and Tualatin do or do not have 512Kb Cache. Don't make a statement you can't back up!
http://www.a1-electronics.co.uk/Intel_Section/CPUs/Tualatinetc_reviews.html
That link says not 512K in Tualitin. They do mention that mobile processors > 1GHz will have it, but I scoured Intel's web pages and couldn't find anything.
http://www.a1-electronics.co.uk/Intel_Section/CPUs/Tualatinetc_reviews.html
That link says not 512K in Tualitin. They do mention that mobile processors > 1GHz will have it, but I scoured Intel's web pages and couldn't find anything.
What about a PIII Xeon? They have big caches, but an equally big price. I think I'd need to sell my liver and kidneys to buy one of those.
Are you saying I can't live without a liver?
Are you saying I can't live without a liver?
Here it is right from Tom himself:
http://www6.tomshardware.com/cpu/00q4/001012/roadmap-02.html
"Coppermine-T and Tualatin
The most logical step to improve Coppermine is obviously to shrink its die down to a 0.13-micron process. This is exactly what is going to happen, but it is by far not all. The next proper version of Pentium III will use a core with the code name 'Tualatin' and it comes with the following specs:
1.13 / 1.26 GHz at launch date
512kB on-die L2 Cache"
Nuff said.
http://www6.tomshardware.com/cpu/00q4/001012/roadmap-02.html
"Coppermine-T and Tualatin
The most logical step to improve Coppermine is obviously to shrink its die down to a 0.13-micron process. This is exactly what is going to happen, but it is by far not all. The next proper version of Pentium III will use a core with the code name 'Tualatin' and it comes with the following specs:
1.13 / 1.26 GHz at launch date
512kB on-die L2 Cache"
Nuff said.
It's gonna use SDRAM?
I guess we'll finally see how 'beneficial' RAMBUS memory is to the P4. I think it's mostly the 400MHz FSB (like it is with the Athlon).
I guess we'll finally see how 'beneficial' RAMBUS memory is to the P4. I think it's mostly the 400MHz FSB (like it is with the Athlon).
***This message to mordy, the guy that said there was nothing that ran at 400MHz***
RAMBUS PC800 memory actually runs at 400 MHz<P ID="edit"><FONT SIZE=-1><EM>Edited by alanschu on 01/08/01 04:18 PM.</EM></FONT></P>
RAMBUS PC800 memory actually runs at 400 MHz<P ID="edit"><FONT SIZE=-1><EM>Edited by alanschu on 01/08/01 04:18 PM.</EM></FONT></P>
Yes yes I know.
But look at the AMD760 benchmarks versus the KT133A benchmarks.
Performance increase more dependant on higher FSB bandwidth than memory bandwidth.
And I know the FSB isn't truly 400MHz, but it's 400MHz effective. Just like the Athlon bus is 200MHz effective.
But look at the AMD760 benchmarks versus the KT133A benchmarks.
Performance increase more dependant on higher FSB bandwidth than memory bandwidth.
And I know the FSB isn't truly 400MHz, but it's 400MHz effective. Just like the Athlon bus is 200MHz effective.
No. It is not effectivly 400 mhz on the p4 or 200 on the athlon. It is 100 mhz DDR. The DDR does not make the performance go up by x4 or x2, there is barely any differnce. The DDR system vs the non DDR system is about 7% (I think it said that in Tom's review of a DDR ATHLON mobo).
And btw, I love those Intel people. They are so charatable, giving money to those monopolistic intel excutives, which use those revenues to pump out brain washing squads and make dumb commercials featuring blue faced people in jumpsuits.
And btw, I love those Intel people. They are so charatable, giving money to those monopolistic intel excutives, which use those revenues to pump out brain washing squads and make dumb commercials featuring blue faced people in jumpsuits.
Well silly me.
I guess I just get the impression that DDR is twice as SDR based on what we've seen with video card memory.
Why wouldn't a FSB being DDR'd or QDR'd be the same way????
And how did Tom test that the 100MHz DDR for the Athlon was only 7% better than an SDR version? I hope that score doesn't factor in other bottlenecks into the equation, because if it does then we wouldn't know for sure (for example, the P3 1 GHz _is_ twice as fast as the P3 500 MHz, but a system like that isn't going to be 100% faster).
Did Tom adjust the score to other bottlenecks? I would think that if there was something that only used the FSB, it would be twice as fast on a 100MHz DDR FSB than a 100MHz SDR bus.
In fact, you talk about Tom's review of the DDR mobo, in which case you are comparing the MEMORY, not the FSB. The Athlon has _ALWAYS_ had the 100MHz DDR FSB (even while using 100/133 MHz SDR memory).
Now that I look over your message, you were talking completely about the Athlon DDR MOTHERBOARD (which I was not).
BTW, in applications where memory is a huge bottleneck, a DDR RAM system will perform pretty much twice as fast, barring any other bottlenecks of course.
In fact, the more and more I read your argument, the more I see you didn't understand what I had written. I had said a 100MHz DDR FSB is 200 MHz effective. I did NOT say it would give you 2X the performance. Just because something is running at twice the speed, doesn't mean it is running twice as fast, and I never once said that it did.
I guess I just get the impression that DDR is twice as SDR based on what we've seen with video card memory.
Why wouldn't a FSB being DDR'd or QDR'd be the same way????
And how did Tom test that the 100MHz DDR for the Athlon was only 7% better than an SDR version? I hope that score doesn't factor in other bottlenecks into the equation, because if it does then we wouldn't know for sure (for example, the P3 1 GHz _is_ twice as fast as the P3 500 MHz, but a system like that isn't going to be 100% faster).
Did Tom adjust the score to other bottlenecks? I would think that if there was something that only used the FSB, it would be twice as fast on a 100MHz DDR FSB than a 100MHz SDR bus.
In fact, you talk about Tom's review of the DDR mobo, in which case you are comparing the MEMORY, not the FSB. The Athlon has _ALWAYS_ had the 100MHz DDR FSB (even while using 100/133 MHz SDR memory).
Now that I look over your message, you were talking completely about the Athlon DDR MOTHERBOARD (which I was not).
BTW, in applications where memory is a huge bottleneck, a DDR RAM system will perform pretty much twice as fast, barring any other bottlenecks of course.
In fact, the more and more I read your argument, the more I see you didn't understand what I had written. I had said a 100MHz DDR FSB is 200 MHz effective. I did NOT say it would give you 2X the performance. Just because something is running at twice the speed, doesn't mean it is running twice as fast, and I never once said that it did.
yoda271828
Distinguished
- Dec 31, 2007
- 527
- 0
- 18,980
Your absolutely right. DDR-SDRAM has twice the bandwidth that SDR-SDRAM has (2.1GB/s vs. 1.064 GB/s). Doubled bandwidth dosn't relate directly to double the performance because it is only one component. Every component in a computer would need to run at twice its current speed to double system performance. Doubling CPU clock speed won't even double system performance.
yoda271828
Distinguished
- Dec 31, 2007
- 527
- 0
- 18,980
Exactly.
So I maintain my statement that the P4 FSB is 400MHz effective, and the Athlon FSB is 200 MHz (and now 266MHz too) effective.
And....
I continue to feel that any performance benefits the P4 has comes from its high FSB bandwidth, and not the RAMBUS modules like Intel/RAMBUS like to insist. The Via KT133A has shown that the majority of the AMD760 DDR chipset speed increase has come from the FSB, not the memory.
Now that that is all straightened out...
So I maintain my statement that the P4 FSB is 400MHz effective, and the Athlon FSB is 200 MHz (and now 266MHz too) effective.
And....
I continue to feel that any performance benefits the P4 has comes from its high FSB bandwidth, and not the RAMBUS modules like Intel/RAMBUS like to insist. The Via KT133A has shown that the majority of the AMD760 DDR chipset speed increase has come from the FSB, not the memory.
Now that that is all straightened out...
"Heck in 10 years or less --- MRAM --- will be out i hear its the fastest ram ever thank you ibm"
Well, I'd like to think that any new RAM would be the fastest RAM ever. Although Rambus seems to have screwed that up
Well, I'd like to think that any new RAM would be the fastest RAM ever. Although Rambus seems to have screwed that up
---------------------------------------------------------
"Well, I'd like to think that any new RAM would be the fastest RAM ever. Although Rambus seems to have screwed that up "
---------------------------------------------------------
WELL - MRAM - is not a form of DRAM, sdram, ddr-sdram, rdram, or even SRAM. Its a new standard.
FINALLY almost out of the old age of DRAM i welcome -MRAM- with open arms.
ABIT BP6 + ATI RADEON 32MB DDR RULE!!!!!!!!!!!!!!!!
"Well, I'd like to think that any new RAM would be the fastest RAM ever. Although Rambus seems to have screwed that up "
---------------------------------------------------------
WELL - MRAM - is not a form of DRAM, sdram, ddr-sdram, rdram, or even SRAM. Its a new standard.
FINALLY almost out of the old age of DRAM i welcome -MRAM- with open arms.
ABIT BP6 + ATI RADEON 32MB DDR RULE!!!!!!!!!!!!!!!!
TRENDING THREADS
-
News Windows 11 will reportedly display a watermark if your PC does not support AI requirements- Started by Admin
- Replies: 19
-
-
-
Discussion What's your favourite video game you've been playing?- Started by amdfangirl
- Replies: 3K
Facebook
Twitter
Instagram