 |
|
mrmitchell
|
|
Maximum PC's "Dream Machine 2003" had two 250GB Western Digital Caviar SEs (7200rpm) in a RAID setup and two 36GB Western Digital Raptors (10000rpm) in another one. Wuss.
|
|
|  |
 |
|  |
 |
|
Asher
|
 |
Calgary, Alberta
Nov 1999 time: 22:30
|
|
quote: Originally posted by Urban Ranger
Who knows what's in store? Certainly not you. |
I certainly would know more about it than you would.
1) Athlon 64 is entering very restricted production in 2003 (something along the lines of 300,000 CPUs TOTAL, before Dec 31st), before cranking it up with the 90nm Athlon 64s in 2004 to the millions. Hardly a threat to Intel.
2) Athlon 64's preliminary benchmarks put it on par with a 6-month old Pentium 4 in 32-bit speed.
3) Intel's pricing roadmap, publically available agrees with me.
So maybe I do know what's in store, huh? It not only makes sense logically and economically, it makes sense because Intel has had their pricing roadmaps public for that time period.
quote: Most people know that a 64-bit processor runs 32-bit applications faster than a 32-bit processor. |
BULLSHIT.
64-bit code takes more memory, which takes more cache, which takes more memory, which takes more memory bandwidth. A 64-bit CPU is not necessarily faster than a 32-bit CPU, until you use the special features of a 64-bit CPU.
Even a novice computer science student would know that MHz != speed, and 64-bit is not always faster than 32-bit.
This right here shows your pathetic qualifications to talk about computer hardware. 
quote: Centrino = AMD 64?  |
Are you really this dumb?
I said compare it to the Centrino. Athlon 64 = modified Athlon = 100+ errata.
Centrino = new core and design = 4 errata, all eliminated before production.
I wonder why it is people think Intel to be more reliable?
|
|
|  |
 |
|
Thue
|
|
Copenhagen, Denmark
Jan 1970 time: 06:30
|
|
More registers are still an advantage. They cut down on the register juggling the compiler would otherwise have to do. This cuts down on the number of instructions executed, the code size, and the memory accesses. That is better than internal registers.
But yes, hyperthreading is cool; I am surpriced that it only gives a 15% speed improvement. I would have expected (without too much knowledge to base it on) a good implementation to easily give at least 50% extra.
Regardless; I will be not be buying a new computer untill after the Athlon64 is out. There must surely be price cuts comming...
Last edited by Thue on 01-09-2003 at 23:49
|
|
|  |
 |
|  |
 |
|  |
 |
|
Urban Ranger
|
|
Apolyton Duke of Off-Topic
|
|
quote: Originally posted by Asher
1) Athlon 64 is entering very restricted production in 2003 (something along the lines of 300,000 CPUs TOTAL, before Dec 31st), before cranking it up with the 90nm Athlon 64s in 2004 to the millions. Hardly a threat to Intel. |
Sure, AMD wants to reserve production facilities for Opteron. Since Opteron is not exactly selling like hotcakes, nothing prevents them from switching production in their new fab to AMD 64.
quote: Originally posted by Asher
2) Athlon 64's preliminary benchmarks put it on par with a 6-month old Pentium 4 in 32-bit speed. |

quote: Originally posted by Asher
3) Intel's pricing roadmap, publically available agrees with me. |
That's right, like Intel hasn't done any surprise price slashes before.
quote: Originally posted by Asher
So maybe I do know what's in store, huh? It not only makes sense logically and economically, it makes sense because Intel has had their pricing roadmaps public for that time period. |
Your logic is as sound as a dried and twisted twig.
quote: Originally posted by Asher
BULLSHIT.
64-bit code takes more memory, which takes more cache, which takes more memory, which takes more memory bandwidth. A 64-bit CPU is not necessarily faster than a 32-bit CPU, until you use the special features of a 64-bit CPU. |
What on earth are you talking about? This does not seem like any sort of counter to my argument, at all.
quote: Originally posted by Asher
Even a novice computer science student would know that MHz != speed, and 64-bit is not always faster than 32-bit. |
Preliminary testing shows that a 1.8GHz runs 32-bit legacy applications faster than P4 3.0C. Your point here is?
quote: Originally posted by Asher
This right here shows your pathetic qualifications to talk about computer hardware.  |
Poor showing, Glonkie. All hot air and no substance.
quote: Originally posted by Asher
Are you really this dumb?
I said compare it to the Centrino. Athlon 64 = modified Athlon = 100+ errata.
Centrino = new core and design = 4 errata, all eliminated before production. |
My point was, of course, your comparison had failed.
If you couldn't figure it out, you really should look for a career in some other field, like basket weaving.
|
|
|  |
 |
|
Asher
|
 |
Calgary, Alberta
Nov 1999 time: 22:30
|
|
quote: Originally posted by Urban Ranger
Sure, AMD wants to reserve production facilities for Opteron. Since Opteron is not exactly selling like hotcakes, nothing prevents them from switching production in their new fab to AMD 64.

That's right, like Intel hasn't done any surprise price slashes before.
Your logic is as sound as a dried and twisted twig. |
Do you understand how weak this is?
1) You just admitted Athlon 64 will be produced in small quantities this year. AMD's marketshare is rather pathetic, and even now Intel's prices are very competitive with AMD's, and Intel's pricing roadmap doesn't have a pricecut til late October. But you're saying this could all change pretty quickly -- and why would it? There's no point, the Athlon 64, at least in 2003, is nowhere near to being a treat to Intel when it's going to be quite pricey. The Athlon 64 FXs aren't going to be out til 2004, anyway.
quote: What on earth are you talking about? This does not seem like any sort of counter to my argument, at all. |
You said a 64-bit processor is faster than a 32-bit processor for 32-bit code. That's patently wrong, stupid, and just laughable.
quote: Preliminary testing shows that a 1.8GHz runs 32-bit legacy applications faster than P4 3.0C. Your point here is? |
http://www.xbitlabs.com/articles/cp...y/athlon64.html
Athlon 64 2800+ vs Pentium 4 2.8GHz









I think my point is made. The Athlon 64 isn't really even competitive with the Northwood Pentium 4s, let alone Prescott. Intel's not worrying about the Athlon 64 for 2003, it's in limited production and the performance is rather shite. The Athlon 64 FX may change that, but that's in 2004.
quote: Poor showing, Glonkie. All hot air and no substance. |
It's the other way around, I provided three points which are based in reality. You simply said "yeah, but that could change!" then said my posting is all hot air and no substance. 
quote: If you couldn't figure it out, you really should look for a career in some other field, like basket weaving. |
Speak for yourself man, this is the latest in a long string of computer threads where it looks like you've regressed in knowledge substantially.
quote: Why? Hyperthreading is overly hyped. Unless you have applications specifically written to take advantage of parallel processing, hyperthreading hardly helps at all. |
Or unless you ever multitask, which I'm sure most people do.
Not to mention how more and more games are being HT-optimized or at least written for SMP.
|
|
|  |
All times are GMT. The time now is 05:30. Apolyton Time is 00:30. |
top of page
|
| archivepost |
|
Forum Rules:
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
|
HTML code is ON
vB code is ON
Smilies are ON
[IMG] code is ON
|
|
|
|
|
|