Newest P4 Review

[Kodiak]

...uhm... OK... what have we learned from the latest review of Pentium4?
Amd is good, but so is P4? It's a complex issue? Things aren't easy? Nothing's as it seems? Tune in next week for an exciting new episode which will reveal all?

Seriously, what (new) piece of knowledge have we come out with...
The internet test hasn't impressed me at all... I completely agree with Tom's statement that he's not quite sure what is does, what it shows and how reliable it is. Hence and therefore, why is it included in the first place?

Personally, while the first three articles helped shape my notion of the new processors, and the forums have had some excellent posts (and some silly ones, the new review seemed like one of those "15 minutes I'll never get back" kinda things...
Opinions?

 

[cipher]

i agree, the internet benchmark seems totally useless and nonsensical IMO

 

[Guest]

This is one of the stupidest things I've heard of in a while. Check out what Tom says:

"This software is what we use all day on the web however, regardless if we despise the missing Athlon-optimizations or not. Adobe Acrobat, Quicktime, Internet Explorer, Macromedia Flash, ... are absolutely common applications. Therefore I would say that we have to respect the high scores of Pentium 4...."

OK, how many Athlon 1.2 Ghz owners out there feel like their 1.2 Ghz CPU is a little sluggish when it comes to browsing one of those "tricky" and resource-intensive web pages with java & flash? I'm sure that a 1.2 Ghz Athlon just isn't quite enogh for IE 5.5; better upgrade now so you don't have to wait 5 minutes for that Quicktime file to decode.... lmao

What IDIOT came up with using these apps for a benchmark? Why didn't they throw in "Notepad" and "Calculator" and some other doozies like "Minesweeper"?

Heh, Tom also said he saw the P4 as more a "life-style emblem" than anything else. I can see those slick guys now: "Loser! My 1.7 Ghz Pentium 4 can run minesweeper at 8,973,421 fps while your crappy athlon only gets 8,973,419!"

OK, I'll lay off the sarcasm. But what was Tom thinking? I mean, how could he even get so far as running the tests without realizing how stupid this is, never mind posting it to the web? If you want to run Flash and Acrobat Reader, go get a $400 e-machines at Kmart w/ a cel 366. P4 1.7 Ghz vs. Athlon 1.466 Ghz? It's just so ridiculous.

STAY TUNED FOR PENTIUM 4 REVIEW #6! THOSE CRAFTY FOLKS AT AMD ARE FINE-TUNING THEIR 3DNOW! OPTIMIZED NOTEPAD, CALCULATOR, AND MINESWEEPER BENCHMARK WHICH, WE'RE TOLD, SHOWS THE ATHLON AT TWICE THE PERFORMANCE OF AN EQUIVALENTLY-CLOCKED PENTIUM 4!!!

 

[Guest]

I agree with your remarks. Many of the tests made today
are not informational because they use applications that
never showed a performance problem since Pentium-MMX.

I am starting to wonder if Quake 3 is still an informational benchmark, when FPS scores hit the 200 mark (!!). (Q3, however, is highly optimized and does reveal
the limits of the processor)

The most important reason that benchmarks should
be really tough (such as the ones in SPEC2000 or similar)
is that when applications are not performance critical
(most office interactive applications are NOT) they never
get optimized properly. Developers in these areas aim
towards a rich feature set and not towards leaner and faster
code. When code is bloated (and speed is sufficient for
interactive use!) benchmarking is pointless! Users don't
care, developer's don't care, why should we use that for
benchmark?

Another thing I would like to mention, by the way, is the
impossible Intel crap behing this NetSomething feature. Even if you have a Pentium Classic 100 you can easily saturate a 10MBps ethernet, let alone a 56K modem. Maybe
a quad server hooked on a 100MBps line needs a faster
processor, but I cannot seriously believe that Pentium 4
makes the internet go faster. Such crap should not be supported by benchmarks such as WebMark.

Petros

PS The information given by WebMark is either (a) useless
if it refers to a client hooked up by a local 100 MBps connection (because nobody has one at home ;-)) or
(b) nonsensical because if you run it from a modem you
will notice that 99 % of the delay is attributed to the modem. (like playing Quake 3 from a 14k modem and blaming
the GeForce 2 Ultra for the lag)

 

[Guest]

Completely right, the lag associated with the internet is due to connection speed and quality much more than processor/fsb speed. Who really cares if a flashing shockwave banner loads .02 seconds faster?

 

[Guest]

All these benchmarks are nice. That's all. Nice.

We should not be as concerned with the CPU's as much as the chipsets that implement them. AMD made a big mistake in letting oems be the primary source for thier chipsets. This opened the door for shoddy workmanship, and incompatabilities. What we need is a review of the different motherboards, but not from a performance perspective. We need a compatability perspective. What cards/memories work well with what boards. This includes video, sound, nic, dvd, and modems. These are all of the most commonly used components. When someone goes to upgrade thier computer with a new mobo, and cpu, it would be nice to know that thier old sound card might have problems with the new chipset, and/or video, or whatever.

It is one thing to review new hardware. this information is good. It is another to review how new hardware interacts with the rest of the system.

I see a lot of problems on this site, and it seems that most are "I upgraded my mobo, and cpu, but now I can't get winxx working.". Most people do not know that thier incompatabilities are caused by the mobo, not the cpu. All they know is that they upgraded to AMD, and it didn't work, or it was such a hassle that they would never recomend AMD to a friend. This is what gives AMD a bad name (unjustly).

Someone should write a book "Chipsets for dummies". I would buy it.

 

[Mordy]

So many people talking about Tom beeing biased
so he had to write an article that make p4 look good (-:

 

[FUGGER]

The same can be said for Sysmark2000, I believe that MS word performance is the true standard for benchmark! NOT

Since AMD won at that test and I don't see anyone bitching about those scores, I figured me being a Pro Intel fan and all I should point that out.

P4 kills AMD when it comes to moving data on the BUS. Sure the AMD has a good FPU but if it cant move the data fast enough it still blows goats. SMP *cough* bottleneck

I need to get the ASUS P4 mobo and try some overclocking. I see where people are getting the 1.4Ghz up to 2.1Ghz with the ASUS.

 

[Kodiak]

re: fugger
I'm not bitching about the scores... before P4 came out it was clear that it would have better bandwidth.. duh... (as long as you go straight and don't change direction often).
But we -knew- that... -- before the first review, definitely before the last...
I was just asking what (*new*) have we learned from the 4th article? I'm just wondering, that's all...

 

[slvr_phoenix]

I think that several of you are entirely missing the whole point of benchmarks.

We know that basically since the P166MMX, every newer processor hasn't really done anything for office applications like Word, Excel, etc. If you're using a P4-1.5GHz, a Celeron-500MHz, a T-Bird 1.2GHz, or a P-166MMX you will see most office applications run just peachy-keen.

And the same is true for the internet and internet applications. They just don't require much processing power to run smoothly. So it doesn't really matter how fast your computer is when editing a Word document or playing Minesweeper.

The benchmarks aren't for saying, "Buy a Pentium 4 and your Word documents will spell-check 2 seconds faster and your life will be so improved by that extra time that you will be able to write dissertations in the time it used to take you to tie your shoes." No one really cares about that. It just isn't the point of the benchmarks.

The benchmarks are for saying, "When comparing the processing abilities of an Intel Pentium 4 to an AMD T-Bird, the Pentium 4 processes such-and-such faster. This means that the Pentium4 provides more processing power than a T-Bird for such-and-such type of applications. And in such-and-such applications, the T-Bird has more processing power. So if you want the best performance for such-and-such, buy this. And if you're using it for this-and-that, buy that."

And THAT is what we all want to know. Even games like Quake don't really matter because no one is going to be able to notice the improved frame rate. It'll play smoothly on either chip.

The point is that based on how much faster these OLD applications run, we can make an educated guess on which chip will run NEW software better.

Software that hasn't even been written yet can still be guaged for performance based on how each processor performs. And so we know which processor runs what type of program better.

And THAT'S the whole point of benchmarks. To show which processors have what strengths and what weaknesses.

NOT to show how much faster minesweeper will run on your three thousand dollar computer.

- Sanity is purely based on point-of-view.

 

[slvr_phoenix]

And what new have we learned from the newest P4 review? Well, using Win2K myself, I learned how it actually does make a difference other than just stability.

I myself didn't learn anything new because I already knew the whole point of the review.

But, others who might like to think that the P4 is a piece of crap might learn that the P4 isn't. It's overpriced, just like any Intel chip is, but it's far from crap. In fact, it can prove it's dominance over the best AMD has to offer.

As much as AMD fans hate Intel, they have to admit that while Intel is releasing newer and faster products, AMD is sitting on their butt and watching the grass grow. And Intel's new products are taking away the title of the fastest chip from AMD.

Despite all else, no one can ignore that. So AMD fans, instead of bashing the P4, you should be complaining that AMD isn't doing anything to counter the P4.

So the final count for now is: AMD makes the cheaper chips. Intel makes the faster chips. Which chip is better? It depends purely on how much money you have to spend.

- Sanity is purely based on point-of-view.

 

[Guest]

If you want to run Flash and Acrobat Reader, go get a $400 e-machines at Kmart w/ a cel 366

I have seen Flash animations (with raster image transitions under the vector work) which were jerky on a PII 450...they only ran at ~8fps even though they were designed to run at 30. Course I think a 1200 would hanlde it

 

[Guest]

What this review really, really shows is that the ball is now in the software industry's court. With today's software, processors are by and large fast enough to perform any duty you throw at them, within a reasonable timeframe, be it an AMD or Intel. Who cares? My money goes to the company that gives me the most bang for the buck. At present time that is AMD. I have a lot of software that will keep me happily working with my computer (AMD Athlon 900mhz, 512 MB RAM, 2x 45GB Raid 0) for a lot of years to come. Most of the time the computer sits there waiting for me to react, not the other way around. Now, would I buy a DVD at $20 and produce DIVX copies on CD's, using 5 hours of a $6000 machine's time to produce a copy? Not very likely. Would I make a copy if it only took 1 hour? Not very likely, either. What I would do, however, is work at my computer for one hour and earn enough money to keep the DVD's coming. The software industry has to come up with new products that are so brilliant that I'll gladly erase Photoshop 6.0 or 3DStudioMax R3 before I'll start caring about which the fastest processor really is. Pentium 4 may be the fastest at programs that are yet to materialize, but at the present, with programs as of today, AMD is king.
Imagine buying a computer that has no upgrade potential, as the chipset and socket will change,with RDRam from a company everybody loves to hate, maybe shaving off half an hour at producing an Mpeg4 from a DVD (From 5 to 4 1/2). And at twice the price? Ya' gotta' be kidding!!!!

 

[Kodiak]

Well said

 

[Grizely1]

Still, Don't forget the clock for clock comparisons.

P4 is good, but not that good.

 

[Guest]

You are not entirely correct in asserting that benchmarks
with todays software can be used to predict future
performance with newer software.

It is certainly true that if one uses consistently e.g.
Word and Excel for Windows then one can guess the
"requirements" of future Word and Excel programs by using
todays versions as benchmarks.

The problem, however, is that the WAY we use our computer
(and I am refering to home users and not to business users)
changes and new software is very often implementing new
algorithms that were previously either (a) non-existent or
(b) not feasible on slow computers.

Let me give some examples:
Case (a) : The famous FPU
- Back in the days of the 486 the FPU was considered a
luxury. All fpu tasks ran slow as hell. Intel even
sold its (infamous!) 486SX without FPU (actually
just like the Celeron and its cache, WITH FPU, only
broken). When the Pentium appeared, FPU tasks where
a reality. The first and most important example is
Quake I. Quake I was the first game to actually
REQUIRE a FPU and AMD was seriously damaged: AMD
processors had been optimized for software such as
Word and Excel. They even got their PR (Pentium
Ratings) which where true enough when referring to
business software. In reality, however, everyone,
including Tom's hardware (I am referring to AMD-K6)
would bash them for the bad FPUs because they had
really bad score in Quake.
Noone could have predicted the need for a speedy FPU
back then because that days software meant "integer"
math only software.


Case (b) MP3 arrives (also FPU intensive, btw)
A few years earlier MP3 did not exist. Now it consumes
huge amounts of bandwidth and space.
Same thing happens today with MPEG4.


The point I am trying to make here is that, like SPEC2000,
when you are trying to make a new and stressful benchmark
you should not choose something that runs really well to
see if it will run faster. You should instead choose something that has just become (or will become) available
in the future, because that is what the users will want
and it will define the requirements posed on the processor.

Flask MPEG is a good choice, not because it make the P4 look
good, but because I believe that a lot of people will be
using it in the future and today it barely works (10-20
hours of waiting is really a lot). It is something NEW
and this is what we will want to run on our new processor.

To conclude my (long!) post I should say that I expect
several new algorithms to become significant in the
future, namely:

- video editing and compression (interesting enough, but
barely feasible today)
- voice recognition (everyone has tried it, but it never
worked because the FPU power was not enough to support
voice recognition as a "side" feature of another task
MS will want to launch something like this!)
- Maybe face/fingerprint/retina(?!) recognition (that
would send e-commerce to new heights and many companies
are already working on biometric security)

Such applications (often contained in forward looking
benchmarks like SPEC2000) do pose new strains to a computing
system (CPU + Memory + Graphics etc) and will determine
how we rate our CPUs in everyday use.

As to the future of CPUs, I believe that Intel is correct
in stressing the need for high bandwidth. Current processors
are not easily saturated.

It remains to be seen whether we will see the addition of
bigger and faster caches (perhaps embedded RAM would help),
more intelligent bus utilisation (auto-prefetch is a great
feature!!!) with less cache misses and branch target misses,
a bigger and faster bus or a combination of the three.

Definitely an 128-bit bus should be implemented but it
would hurt current 32-bit performance. (to make it
efficient one would have to request as many as 3 or 4
words for every cycle and in 32-bit processors this is
not possible - not enough units , it would be extremely
fast in FPU math, however, because current FPUs are
capable of using two 64-bit quantities per cycle).

Faster busses are nice but they come with added complexity
and they need new memory technology such as dual channel
RDRAM to feed them. Such technology is expensive.

ANW, here it is 01:30 AM and I should be sleeping. If I
have written anything stupid you should attribute it to
my sleepiness.

Petros