AMD slower ghz?

mgoldb2 · 06-15-2005, 04:53 AM

Quote:

Originally Posted by NeuromancerWGDD'U

Anyway, my point in the whole "Intel not being 64-bit" bit is that Zabarl is comparing the operating frequency of a 64-bit AMD chip, to the operating frequency of an Intel 32-bit chip.

It does not matter if one 32bit and the other 64bit unless you using a 64bit operating system. A sempron is 32bit and it still operates at a lower speed then a celron but the sempron will perform just as good if not better then the celron. In some cases a sempron is better then P4's. It all depends on exactly what two cpus you comparing.

Anouther example 3.0ghz p4 64bit(630) performs almost no different the a 3.0ghz p4 32bit(530).

What am trying to say is being 64 bit has nothing to do with why it performs better then a intel of the same speed.

**The_Other_One** · 06-15-2005, 05:07 AM

AMD's speeds are much better than Intel's(when compared to the same MHz) Man, compare a Celeron 1.6GHz to a Sempron 2800(well, my mobile one runs at 1.6GHz) Heck, I have it underclocked at 800MHz now to save battery and I bet it'd still be most celerons

Probably put up a good fight with some P4's too

**Praetor** · 06-15-2005, 06:18 AM

Quote:

How come the AMD 64 GHZ speed is slower than an Intel?

One could ask "Why is the Intel one faster?" ... the point being the clock speed isnt an issue. For instance, most people buy a car based on a make and model number rather than the number of revs in such a gear at such a speed

Same with computers (i.e., you buy a make and model rather than by a specific clock frrequency)

Quote:

Like my 3700+ san diego is 2.2ghz and isaround 330 bux while a intel 3.4 prescott is around 310 bux

1. More expensive doesnt mean better. Nor does it mean worse
2. The Pentium 4 3.4GHz is, afaik, discontinued in favor of the the Pentium 4 550

Quote:

Does that mean the 64 is slower?

Well it doesnt "mean" that it is slower. The fact of the matter is that the AMD Athlon64 3700 is slower. Why is kind of irrelevant

Quote:

In my opinion the AMDs are faster than the Intels.

And important distinction here is that for specific tasks (namely gaming), AMD processors (and in particular their K8 lineup) is better than intel processors -- not faster ("better" is measured in something like fps or seconds whereas "faster" is measured in Hz)

Quote:

The 3700 means it's equal to a P4 3.7 for example

Thats what most people think and AMD knows it. 3700 is an arbitrary number given to a specific type of chip running at 2.2Ghz

Quote:

AMD chips perform 9 operations per clock cycle

Ok before i see more of this crap... what opcodes execute in 1/9th of a cycle? LIST THEM. Last time i checked ASM for x86 the most efficient ops took place in 1 cycle. So lets drop the BS and talk facts.

Quote:

AMD chips perform 9 operations per clock cycle, whereas Intel chips can only do 6 operation per clock cycle

Even if we buy into this silliness, the 3700+ at 2.2Ghz x 9 = 198000000000 and say a 560 which is the "closest" at 3.6Ghz x 6 = 216000000000... so wth are you talking about?

Quote:

I have one of the new Intel 64-bits "630" and it better be 64-BIT! It's still no where near the performance of an AMD anything bit!

Its a 64bit processor. Of course whether you're running it in 64bit mode depends on th OS involved.

Quote:

I believe Intel has 64-bit processors as well, and now that there are applications which actually take advantage of that, such as Windows 64-bit edition...its starting to be worth it.

Yes Intels 6xx linup of consumer processors and everything to follow will generally be 64bit processors

Quote:

My avatar shall rule the world!!! Actually I got the pic off some news site. I think he's a congressman or something. It still looked funny

Lets stay on topic please, avatar discussion can happen in the General chat

Quote:

Ah, I had read occasional spatterings of internet reviews and the sort, and kept hearing about how Intel had yet to go to 64-bit. I suppose then, that those were outdated writings. Ah well, I don't pay too much attention to what Intel does, after all, it's been a long time since they've done anything innovative or earth-shattering.

Hmmm funny how Windows XP 64bit edition was designed for Intel platforms and then awhile later Windows XP x64 came out for AMD processors..... I'd say Intel's been in the 64bit market a lot longer than you give them credit for

Quote:

Anyway, my point in the whole "Intel not being 64-bit" bit is that Zabarl is comparing the operating frequency of a 64-bit AMD chip, to the operating frequency of an Intel 32-bit chip.

But the external addressing size of the processor doesnt matter

Quote:

Yeah, well, i'm not paying anymore attention to Intel anymore. They promised me speed, and they gave me nothing but a bigger price tag... If I didn't spend 230!!! freaking dollars on the CPU alone I would definately go back to AMD.

Intel processors are better suited for certain tasks and AMD processors suited for other tasks. Simple as that. Sure you can argue a Hummer (a real one, not them silly ones you see civilians driving) to outpace a Ferrari you'll be sadly mistaken but on the flipside, no Ferarri is gonna survive the roughzones like a Humnmer

Quote:

AMD's speeds are much better than Intel's(when compared to the same MHz)

Its been a long time since they've been the same though hehe ... ever since launch of Netburst

NeuromancerWGDD'U · 06-15-2005, 02:11 PM

Quote:

Originally Posted by Praetor

Ok before i see more of this crap... what opcodes execute in 1/9th of a cycle? LIST THEM. Last time i checked ASM for x86 the most efficient ops took place in 1 cycle. So lets drop the BS and talk facts.

Overall, AMD Athlon XP processors are able to perform 9 operations per clock cycle while Intel can only manage 6.

SOURCE: http://www.pcmech.com/show/processors/715/

AMD cpus (athlon 32 bit) perform 9 operations (units of actual work)
Intel cpus (P2-3, older P4 and newer P4 while not in a hyperthreading state) only do 6 operations per clock cycle...

so you do the math...

*opcc = operations per clock cycle
*ops = operations per second
EDITED FOR ACCURACY
(AMD) 9 opcc multiplied by 2000 (2ghz) x 1,000,000 = 18,000,000,000 ops (eighteen billion)
(INTEL) 6 opcc multiplied by 3000 (3ghz) x 1,000,000 = 18,000,000,000 ops

realistically, the exact same amount of real world work, regardless of overrated megaherts hype....

SOURCE: http://forums.pimprig.com/archive/topic/43233.html

The athlon does 9 operations per cycle ... how many does the p4 do??? Yes that's right ... only 6. Pretty poor show really.

SOURCE: http://www.geek.com/news/geeknews/20...1114008864.htm

AMD Athlon CPU make up its disadvantage in memory bandwidth by providing three Full x86 decoders (while Pentium 4 has only 1) and performing 9 operations per clock cycle (while the Pentium 4 has 4). Consider the much more expense on the latter system, we still believe the former one is worth to report here.

SOURCE: http://64.233.167.104/search?q=cache...ck+cycle&hl=en

Also, the Athlons perform 9 operations per clock cycle, compared to 6 for the P4 (non HT).

SOURCE: http://www.webhostingtalk.com/archiv.../103451-1.html

AMD Athlon XP 2800+ (2.083Ghz) (Inc fan/heatsink) (OEM)

# Frequency 2.083 GHz
# Cache Size: L1 - 128KB and L2 - 512KB
# Die Size: 128mm2
# Transistor count: 37.5 million
# Infrastructure: Socket A
# QuantiSpeed™ Architecture
# 9 Operations per clock cycle

SOURCE: http://www.pluscorp.com.au/Product.a...=2&ProdID=2387

Infrastructure: Socket A
QuantiSpeed™ Architecture
9 Operations per clock cycle

SOURCE: http://dljsystem.com/detailsCPU.asp?productID=1328

Intel's do 6 operations per clock, AMD's do 9 operations per clock.

Simply put, AMD's do more calculations, and have less bottlenecks when gaming.

I could write you a book about it, but I'd just be quoting AMD...and even I would get lost in the technical explanation.

SOURCE: http://www.overclock.net/archive/index.php/t-6210.html

So AMDs or Intels?
Let me out an example of the XP 3200. It only goes up to 2.2ghz. On the other hand the P4 goes up to 3.2ghz and over. *ok more ghz, im happy*.
Unforuntely, no. the Intel does only 6 floating points per cycle meaning that it can only carry out 6 operations per cycle. The AMD does 3 more so it can do 9 points.

Maths Lession:
3.2ghz x 6 = 19.2 (Intel)
2.2ghz x 9 = 19.8 (AMD)

SOURCE: http://www.kirupa.com/forum/showpost...88&postcount=2

OK, Praetor, now how precisely can you call this bullshit? Obviously something's amiss, because I couldn't find one shred of credible information that even remotely refuted my statement. In fact, it would seem that there is an overwhelming tendency to agree with my statements. By all means, you may ask for my sources of information, but before you call my words "crap" and/or "BS," at least confirm that they are "crap," and/or "BS."

NeuromancerWGDD'U · 06-15-2005, 02:23 PM

Quote:

Originally Posted by Praetor

Hmmm funny how Windows XP 64bit edition was designed for Intel platforms and then awhile later Windows XP x64 came out for AMD processors..... I'd say Intel's been in the 64bit market a lot longer than you give them credit for

Yeah, kinda embarrassed about that, I really need to pay attention to both sides of the field there... anyway, you must also keep in mind that Intel owns a very large percentage of the commercial processor market, and it would be very advantageous for Microsoft to initially release x64 for Intel's 64-bit processors versus AMD's 64-bit processors.

**Praetor** · 06-15-2005, 11:24 PM

Quote:

Originally Posted by http://www.pcmech.com/show/processors/715/

AMD Athlon XP’s have 3 X86 decoders, 3 floating-point pipelines, and 3 integer pipelines. This is compared with Intel’s Pentium 4, which has only one X86 decoder, 2 floating-point pipelines, and 1 more integer pipeline than AMD’s Athlon. This leads to AMD being able to decode more instructions than Intel at the same time, and being able to perform floating-point operations quicker than Intel. Overall, AMD Athlon XP processors are able to perform 9 operations per clock cycle while Intel can only manage 6. It doesn’t sound like much, but in processors every operation is crucial. This is why I said AMD are more about getting more done per clock cycle in my AMD processor buying guide.

Unfortunately thats quite inconsequential if not partially incorrect:

You cannot break an opcode -- even when an interrupt is called, the currently executing opcode will complete before the interrupt is dealt with. This holds true of 8086 and 68xx and 68K assembly. I dont know specifically about other ASM platforms but it would make sense that they would not be able to split an opcode.
Do make sure you are aware of the distinction between an operation and an instruction as they are not neccesarily the same.
Now what the comment in the quote was referencing was about the superscalar architecture which was introduced to deal with the idea of "why are we limited to a maximum of 1 operation per cycle?" -- something you didnt mention. Regardless, superscalar architecture simply means "a crapload of extra silicon was added to the chip in the form of decoders and execution units etc. The funky thing about this is that, like all neato inventions ... there is a catch (you didnt think it'd be that easy did you hehe) and the catch is that all this fancy hardware only kicks in if the multiple instructions in the prefetch queue can be executed independently

Getting back to the point Im making about execution times:

The x86 processor doesnt an opcode within a single cycle. Consider something simple and stupidly common like mov ax, bx and we'll break it down in a bit more detail
- Fetch instruction. 1 cycle
- Update the IP. 1 cycle
- Decode instruction. 1 cycle
- If needed, fetch a 2-byte operand. 0-2 cycles
- If such an operand was required, update the IP again. 0-1 cycles.
- Translate the address of that operand if needed. 0-2 cycles
- Fetch that operand now that we know its address. 0-3 cycles
- Save the value to the destination register. 1 cycle
The total here is anywhere from 5-11 cycles for a stinkin mov op This is a good analysis of the MOV instruction

That being said, only one serious correction needed to that first link you provided:

Quote:

Overall, AMD Athlon XP processors are able to perform 9 operations per clock cycle while Intel can only manage 6

While it may just be a case of symantics but an important one. When talking about executions or efficiencies or such, the unit being dealt with is an "operation" or "an opcode" (when we breakdown an opcode, sometimes that's called an 'instruction' but that is explicitly noted). Any other time 'instruction' is used, it's synonymous with 'opcode'. What that quote should have said was "AMD Athlon XP processors are able to prefetch 3 op-instructions concurrently and deal with three such FP and INT instruction streams and the Intel Pentium 4 is only able to prefect a single instruction and deal with two FP and a single INT instruction stream".

That being said, perhaps you should have a look at page two of the link you gave me. Contrary to what you try to pass as the number of decoders and execution units being the defining characteristic of a processors performance -- it's not. (That and the fact that not all instructions are able to be executed independently as noted above -- then again thats what OOOE Optimization is for hehe)

-----

Quote:

(AMD) 9 opcc multiplied by 2000 (2ghz) x 1,000,000 = 18,000,000,000 ops (eighteen billion)
(INTEL) 6 opcc multiplied by 3000 (3ghz) x 1,000,000 = 18,000,000,000 ops
realistically, the exact same amount of real world work, regardless of overrated megaherts hype....

If you're trying to suggest that either of those processors will execute 18billion opcodes per second, then let me suggest to you that the year is 2005. Not 3005 hehe

----

Hehe the link gave me quite the chuckle .... its a collection of both AMD and Intel fanboys at a convention. One quote that stood out hehe

Quote:

Originally Posted by amdtel

The worst thing that can happen to AMD is that if they degrade their product by putting more pipelines.....if they do it in hammer they are screwed

<sarcasm> Well I guess they're screwed arent they? </sarcasm>

-----

Quote:

AMD Athlon CPU make up its disadvantage in memory bandwidth by providing three Full x86 decoders (while Pentium 4 has only 1) and performing 9 operations per clock cycle (while the Pentium 4 has 4). Consider the much more expense on the latter system, we still believe the former one is worth to report here.

While true from a hardware level (again the wording is quite wrong -- I wish my procs could be that effective), they failed to note that the reason any old-generation Pentium4 loses out to AthlonXP and Pentium IIIs is because their clock speeds are not scaled enough to counter the ineffectiveness of the individual cycle

----

Quote:

Also, the Athlons perform 9 operations per clock cycle, compared to 6 for the P4 (non HT).

Again with the wording .... just because you find forum comments about the number of "operations" being done doesnt mean thats actually how it is. Because if it was, I'd settle for that 18billion-opcodes/sec setup you 'calculated' earlier. Yer a bright kid I think and i'll assume you get my point about the wording

----

Quote:

Simply put, AMD's do more calculations, and have less bottlenecks when gaming.
I could write you a book about it, but I'd just be quoting AMD...and even I would get lost in the technical explanation

.
Well a few things about that:
- The distinction for gaming has only really shone through with the K8 based processors and their on-die memory controllers ... having a memory controller that operates in synch with the cpu rather than the NB is hell of a boon and that is what gives AMD processors their killer advantage over Intel silicon in the gaming market
- Just because a processor is suited for gaming does not mean it's suited for everything. Although this is beginninng to change (although only with the top end processors), real work is still an Intel silicon job

Quote:

So AMDs or Intels?
Let me out an example of the XP 3200. It only goes up to 2.2ghz. On the other hand the P4 goes up to 3.2ghz and over. *ok more ghz, im happy*.
Unforuntely, no. the Intel does only 6 floating points per cycle meaning that it can only carry out 6 operations per cycle. The AMD does 3 more so it can do 9 points.

Well in that case with those two processors, explain to me why
- Intel processors dominate the photoshop sector? To a degree that's an INT bount application and since the AMD proc has supposedly 3x the INT capacity ... why does the Intel one win out?
- Intel processors dominate the Premier sector? Thats a FP arena and the AMD chip there should have an advantage since it has an extra FP pipeline?
- Some other supposedly FP bound stuff ends up in Intel land?
- And in animation, a FP bound application for CPUs, ends up in Intel land?

.... continues

**Praetor** · 06-15-2005, 11:24 PM

... continued

Quote:

OK, Praetor, now how precisely can you call this bullshit? Obviously something's amiss, because I couldn't find one shred of credible information that even remotely refuted my statement. In fact, it would seem that there is an overwhelming tendency to agree with my statements. By all means, you may ask for my sources of information, but before you call my words "crap" and/or "BS," at least confirm that they are "crap," and/or "BS."

Well the simple fact that you seem to have confused the meaning of "operation" gives things away. Furthermore, just because you found some links here and there (mostly forum ones it seems) that corroberate doesnt mean much when you dont apply the facts being conveyed with a logical sense of reasoning. Sure the AthlonXPs can have 3 FP units ... I didnt even bother verifying that that (although didnt need to either) .... but just because they have three such units does not mean they are better processors .... and on the converse, just because intel doesnt have 3 such units does not make it a better processor. Since Ive dealt with ASM and I know for damn sure opcodes dont execute in fractions of a cycle I countered by asking for specific opcodes that may execute in 1/6 or 1/9 of a cycle (since, not being an ASM guru I dont know the entire opcode layouts off the top of my head). That was an opportunity to solidify your case. I see no opcodes.

Regarding the "applying the facts" comment, sure AMD can have 9 "whatever" per cycle and Intel only have 6 "whatevers" per cycle, but two things should come to mind:
1. Why the hell are all these processors so slow??? Sure we know its CISC technology so yer looking at a 300% bloat in opcode length but at say 18billion opcodes/sec -- it shouldnt matter
2. The fact that AMD has more "whatever" does not mean it is a better processor. And never will. The fact that Intels have fewer does not mean it is a worse processor and never will.

I think an overwhelming majority of informed users will concur that to say "AMD makes the best processors ever and all Intel Processors suck" (even if we limit the scope to current and last-gen processors) is an excessively broad and closeminded view. There is a reason why Intel and AMD exist and why there are so many threads on the internet about "Intel or AMD" -- thats because its not hands down clear.

----

Quote:

Yeah, kinda embarrassed about that, I really need to pay attention to both sides of the field there... anyway, you must also keep in mind that Intel owns a very large percentage of the commercial processor market, and it would be very advantageous for Microsoft to initially release x64 for Intel's 64-bit processors versus AMD's 64-bit processors.

1. Windows XP x64 is not [originally] for Intel systems but rather for AMD's K8 lineup. I think you have Windows XP x64 and Windows XP 64bit Edition mixed up?
2. The fact that Intel has the lions share of the consumer market was not why Microsoft released Windows XP 64bit edition -- t'is cuz Intel had their 64bit stuff up and running a lot sooner than AMD

----

Oh and on a last note:

Quote:

Geforce FX 5700LE @ stock 250/500 core/memory with 256-bit memory interface, **drool**, ever so unneccesary...

That videocard doesnt have a 256bit memory interface.

98vert6spd · 06-15-2005, 11:42 PM

Quote:

Originally Posted by Praetor

... continued

Well the simple fact that you seem to have confused the meaning of "operation" gives things away. Furthermore, just because you found some links here and there (mostly forum ones it seems) that corroberate doesnt mean much when you dont apply the facts being conveyed with a logical sense of reasoning. Sure the AthlonXPs can have 3 FP units ... I didnt even bother verifying that that (although didnt need to either) .... but just because they have three such units does not mean they are better processors .... and on the converse, just because intel doesnt have 3 such units does not make it a better processor. Since Ive dealt with ASM and I know for damn sure opcodes dont execute in fractions of a cycle I countered by asking for specific opcodes that may execute in 1/6 or 1/9 of a cycle (since, not being an ASM guru I dont know the entire opcode layouts off the top of my head). That was an opportunity to solidify your case. I see no opcodes.

Regarding the "applying the facts" comment, sure AMD can have 9 "whatever" per cycle and Intel only have 6 "whatevers" per cycle, but two things should come to mind:
1. Why the hell are all these processors so slow??? Sure we know its CISC technology so yer looking at a 300% bloat in opcode length but at say 18billion opcodes/sec -- it shouldnt matter
2. The fact that AMD has more "whatever" does not mean it is a better processor. And never will. The fact that Intels have fewer does not mean it is a worse processor and never will.

I think an overwhelming majority of informed users will concur that to say "AMD makes the best processors ever and all Intel Processors suck" (even if we limit the scope to current and last-gen processors) is an excessively broad and closeminded view. There is a reason why Intel and AMD exist and why there are so many threads on the internet about "Intel or AMD" -- thats because its not hands down clear.

----

1. Windows XP x64 is not [originally] for Intel systems but rather for AMD's K8 lineup. I think you have Windows XP x64 and Windows XP 64bit Edition mixed up?
2. The fact that Intel has the lions share of the consumer market was not why Microsoft released Windows XP 64bit edition -- t'is cuz Intel had their 64bit stuff up and running a lot sooner than AMD

----

Oh and on a last note:

That videocard doesnt have a 256bit memory interface.

0WNED

**Praetor** · 06-16-2005, 12:08 AM

Well it wasnt meant as an "attack" per se.... it's just that yes he makes a point that "Hz is not the bottom line" but then again neither is "number of execution units"

kof2000 · 06-16-2005, 12:12 AM

i think you're forgetting about the 630 also be HT? whereas the amd they release the x2 for that.

06-15-2005, 05:07 AM	#12 (permalink)
The_Other_One Super Moderator Join Date: Mar 2005 Location: Lexington, NC Age: 23 Posts: 12,185	AMD's speeds are much better than Intel's(when compared to the same MHz) Man, compare a Celeron 1.6GHz to a Sempron 2800(well, my mobile one runs at 1.6GHz) Heck, I have it underclocked at 800MHz now to save battery and I bet it'd still be most celerons Probably put up a good fight with some P4's too __________________ Urban De-Development * Athlon X2 4200 \| 2048M RAM \| 160G HD \| 7600GT * C2D T5250 \| 2048M RAM \| 160G HD \| Intel x3100 * G4 1.33GHz \| 512M RAM \| 40G HD \| Radeon 9200

06-16-2005, 12:08 AM	#19 (permalink)
Praetor Administrator Join Date: Jul 2004 Location: Canada Age: 24 Posts: 19,900	Well it wasnt meant as an "attack" per se.... it's just that yes he makes a point that "Hz is not the bottom line" but then again neither is "number of execution units" __________________ ASUS P5K Premium WiFi-AP, Q6600@3.7 / ASUS P5ND, E6400@3.8 4GB OCz Platinum XTC 8500 / 4GB CorsairXMS2 6400 5x500GB Seagate 7200.10 / 2x500 Seagate 7200.10 OCz 8800GTX 768MB @ 630/800 / 2x Galaxy 8800GT SLI

06-16-2005, 12:12 AM	#20 (permalink)
kof2000 Diamond Member Join Date: Aug 2004 Location: Oahu Age: 26 Posts: 6,273	i think you're forgetting about the 630 also be HT? whereas the amd they release the x2 for that. __________________ QX9650@4.2 swiftech \|9600@3.0 9700 nt 780i \| 790fx 8gb ddr1200 \| 8gb ddr1200 2x150gb 3x1tb \|2x250gb 3x8800gtx \|3x2900xt 1gb 2x20xdvdrw \|2x20xdvdrw 15in1 card reader \|15in1 card reader ng15/g9 \|ng15/g9 1200w psu \|1200w psu 2x28" LCD \|2x28" LCD

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode