> I'm pretty sure that not all instructions are equal (eg, take equal time to execute) therefore I don't see how one processor can be said to do so and so instructions per cycle. <
True, but in those tests I seriously doubt they would test one CPU with the compilation of a HUGE program, as opposed to the compilation of HelloWorld.c
So, to your remark, run the same exact test on each CPU.
The world is waiting. I must leave you now.
No, i'm saying that; for instance an XOR can be performed faster than a MOV.
If you are interested in that level of difference, you can check out a site
where that's their bread and butter, assembly board . Do a search, try
"AMD", "relative processor speeds", "benchmark testing", and such. Their
board works just like this one. Those guys have done and are doing every
kind of benchmark testing you can think of, including cpu comparisons,
on big and small files and programs, exe's produced by different
compilers on the same code, all kinds of stuff.
And, believe it or not, it's a very active and lively board. Much like
this one!
Oh, I almost forgot to mention, some of their discussions about the
benchmarking methods themselves can be, well, rather entertaining,
as well as well-informed. And, you can find links to download various
custom benchmarking programs and do the testing yourself.
globalerr_h
Win98se
Dev-C++ 4
Borland55 with JFE
No, not wrong. In the first place, you don't have the requisite knowledge to even know the difference, or whether or not the article is flawed or not. No one can even compare AMD Athlon's against Pentium's P4s-- they have completely different architectural pipelines. That's why this entire argument is ludicrous. It is a fact that there or _NO_ benchmark programs out there written to accurately evaluate a P4.Originally posted by Ride -or- Die
WRONGGGGGGGGGGGGGGGGGGGGGGGGGGGGG! Did u read that article? if you said yes go back and pay attention this time.
It's like constantly comparing Apples & Oranges. And the fact is, some programs won't run on an AMD because of its intrinsic differences.
Also, someone said AMDs do 9 instructions per cycle, while Intel P4s only do 6-- that's not correct. The number of instructions executed per cycle is _variable_ based on what instructions are used, up to a maximum per cycle count.
It is not the spoon that bends, it is you who bends around the spoon.
>
Also, someone said AMDs do 9 instructions per cycle, while Intel P4s only do 6-- that's not correct. The number of instructions executed per cycle is _variable_ based on what instructions are used, up to a maximum per cycle count.
<
Yes or no Sayeh, under the maximum per cycle count, is the 9 vs 6 comparison true?
I don't know the internal workings of it all like you do Sayeh, even though I would love to come to grips with it all, I will stick to AMD's 2800+ CPU as Intel cannot stomach to offer prices similar to AMD.
Shoot, the MAME group who are emulating arcade video games that require more CPU horsepower(it's all about your CPU nothing else..crappy graphics and sound card work fine for MAME) than is vailable through home PC overclocking are only using 2ghz P4's and AMD's...
I don't know how it all works and from you I'd love to here a Sayeh rant(details) to end this discussion once and for all, so, lets here it.
The world is waiting. I must leave you now.
>>No, not wrong. In the first place, you don't have the requisite knowledge to even know the difference, or whether or not the article is flawed or not.
You don't know what knowledge i have.
>>No one can even compare AMD Athlon's against Pentium's P4s-- they have completely different architectural pipelines. That's why this entire argument is ludicrous.
Actually i can, see its very easy. The fact that AMD is different then intel is they're fault/plroblem, and i don't see why i should limit my opinion because of a "handicap" they have. I don't think the argument is ludicrous, if your going to supply a product in the same field as another company, how you make/provide that product doesn't remove you from comparison, period.
>> It is a fact that there or _NO_ benchmark programs out there written to accurately evaluate a P4.
You might be right on that one, i honestly don't know.
>>And the fact is, some programs won't run on an AMD because of its intrinsic differences.
Another reason to go intel.
>>Also, someone said AMDs do 9 instructions per cycle, while Intel P4s only do 6-- that's not correct. The number of instructions executed per cycle is _variable_ based on what instructions are used, up to a maximum per cycle count.
Actually they were right, and your incorrect.
>>And the fact is, some programs won't run on an AMD because of its intrinsic differences.
And vice versa.
>>Actually i can, see its very easy. The fact that AMD is different then intel is they're fault/plroblem, and i don't see why i should limit my opinion because of a "handicap" they have. I don't think the argument is ludicrous, if your going to supply a product in the same field as another company, how you make/provide that product doesn't remove you from comparison, period.
This isn't as simple as taking two cars and seeing which one will go faster down a road. The processors don't/can't use the same road. Just because AMD makes their chips differently than Intel doesn't give them a handicap; it just makes them different. Period.
>>Actually they were right, and your incorrect.
No, he's correct. Go read the article JoeSixPack posted at the start of this thread.
>>No, he's correct. Go read the article JoeSixPack posted at the start of this thread.
I stand corrected.
howdy,
very simply put.
my dual AMD mp 2000+, Asus A7M266-D box runs circles around any Intel box i know of, and i know of a lot, being a contractor to Intel at Rio Rancho.
this is based on what "I" use my box for, not on a generic users multiple application most of the time use basis.
my point- for "MY" application AMD is the right thing the right way (stolen from Guy Kawasaki).
to sit and try to cast a CPU into a mold based upon everything all of the time is a waste of time.
i buy stereo equipment that sounds good to "ME" i buy computer harware that works good in "MY" application.
benchmarks, Ghz, performance comparisons be damned.
M.R.
Last edited by itld; 11-22-2002 at 09:58 PM.
>
i buy stereo equipment that sounds good to "ME"
i buy computer harware that works good in "MY" application.
benchmarks, Ghz, performance comparisons be damned.
<
..and the truth is told.
> stereo equipment
Mmmm, nice discussion, but, backs up your overall point.
Who's to say what's better?
Insane volume with comfortable, light-weight quality(good average brand)?
Or something over-finessing trying to suck every bit of quality out of the product as possible? (B&W's nautlius speaker which costs $20,000)
It's a matter of opinion.
Some like art - quality
Some believe nothing compares with awesome power - volume
Going on and on about it is absolutely pointless.
Critics are worthless.
The world is waiting. I must leave you now.
Buy what you want and what you're happy with.....it's as simple as that.
Last edited by damonbrinkley; 11-23-2002 at 05:15 PM.
this thread is a waste of db tables.....
man has this post gone on a while, what I really want to know is though is if a pentium and amd each with the same clock speed, have different bus speads then wouldn't it mean that either the amd has to slow a bus to make use of its speed, or intel has a bus which is too fast to be of much use to the processor and thus is wasted. Does amd send more compressed processed data? Is the only reason intell needs the the extra bus spead because its cache is smaller? Oh and vvv your sigs to long.
