Slashdot Mirror

woops, should have had a link to Anandtech

anandtech reviewed 18 units on july 31.

Interesting there is the memory test: they show that stable power gives less memory errors with memtest.

And here is clickable tomshardware:
Toms burns up some power supplies

Read this page from AnandTech's PSU roundup. The only thing that was changed between tests was PSUs, yet over a 6 hour time there was a range of only 1 up to 7 memory errors. Just one possible indication of how clean the supply is. Also there's the added benefit that it probably won't die in a month or two (happened to me, not too pleased when it did)

AnandTech also just had a PSU roundup here. The watt numbers and some of that aren't the best from AnandTech's review (read some criticism of it here), but overal it's a good roundup, especially comparing heat and noise.

AnandTech also just had a PSU roundup here. The watt numbers and some of that aren't the best from AnandTech's review (read some criticism of it here), but overal it's a good roundup, especially comparing heat and noise.

Power Supply Roundup Part 2

Not a Mobile Pentium 4.. a Pentium-M. It was designed specifically for laptop and mobile systems. There's an article about it on Anandtech here.

I agree that the iBook is a great laptop -- I've had mine for 2 years now. But that doesn't mean you should discount the competition so quickly.

Firt of all, the Centrino is a processor (the Pentium M) + built in wireless card. It's not a laptop. That would be like calling the G3 a laptop. Lots of different companies make P-M/Centrino based laptops. The Centrino compares favorably with the P4 in terms of performance, while consuming remarkably little power. As Apple themselves will admit , a single P4 compares pretty good to the new G5. So it would seem crazy to think that a G3 can compare at all to a P-M in terms of performance. It also compares well in terms of battery life.

Centrino is one of Intel's most remarkable accomplishments. It's a processor specially designed for laptops. It would be very hard for Apple to have such a processor, since who could they get to do the R&D for it?

As I mentioned earlier, dozens of different companies make Centrino based laptops. The overall quality varies quite a bit. More often than not, the quality is well below anything Apple would make. But sometimes it is not. Be careful when making blanket statements. If you want to see a really nice Centrino check out IBM's ThinkPad X31. It's 1.19" thick, 3.6 lbs., gets 4.5 hrs of battery life, and while not the fastest Centrino, is plenty fast compared to a G3.

Finally, it's hard to imagine that a whole lot more performance is going to be squeezed out of the G3. The G4 only got up to 1.42 GHz, so it's hard to expect the G3 to get over that.

Do your part for Internet Free Speech. Boycott Toms Hardware Guide.

And if you don't give a rat's ass about Internet Free Speech, boycott Tom's Hardware because they suck. Articles are spread across too many pages simply to create ad revenue, articles are poorly written and researched, the editors often seem to take a cue from Slashdot, and to top it all off THG is hardly impartial. If you want good hardware coverage, get it somewhere else.

1. The cards are nearly identical in speed. In fact, benchmarks done by others (like Tom's Hardware and Anandtech) seem to show the FX5900 edging the Radeon in most tests. (You may be thinking of the 5800, which was, indeed, slower than the Radeon.)
2. The FX5900's that you have seen benchmarked are all running at 450/850. The eVGA version of the FX5900 is clocked at 500/900, which is possible because they put 2ns VRAM on their card. Naturally, this means a 5%-10% performance boost, allowing it to edge out the Radeon in more tests.
3. The best GeForce FX and the best Radeon cost the same at $500 (last I checked, about two weeks ago).
4. The FX 5900 allows far more complex vertex and pixel shaders. Pixel shaders can be 1024 instructions long and may include branches. I think the Radeon's limit is, like, 16 or 32 instructions, with no branches, but don't quote me on that.
5. The FX 5900 runs Doom 3 much faster. I know this isn't relevant now but it's an interesting point. Current games are going to run unbelievably fast on either card, but future games will run faster on the FX 5900.
6. The GeForce FX 5900 fan is not loud. The infamous "dust buster" fan was on the 5800. The 5900 uses a more traditional fan. The only time you can even hear it is when you open a game, and it's really not loud at all. I don't even notice it unless I'm listening for it (yes, I own one).

There really is no clear winner between these two, and they cost the same. So why wouldn't people buy the FX? I prefer to support NVidia because they brought about all the recent great leaps in graphics technology (programmable vertex and pixel shaders, Cg, etc.) whereas ATI hasn't come up with anything particularily impressive.

NVidia is not 3dfx. Don't expect them to die anytime soon.

(I am a professional game programmer. Just thought I'd mention that.)

All I did is point out that the Apple rep didn't mention the Quake benchmark which is *way* off for the P4.

Here are some regular PC scores that bear no resemblance to what Apple quoted

Anandtech

SharkyExtreme

Now this has been widely reported. So I *know* you don't think me a troll just because I was curious why the Apple rep didn't mention these stats.

A 3GHZ P4 (which has been out for a while) scored 385.6 fps. According to Apple's website it is 275. That is a *huge* difference.

Apple Bench

Now since you so clearly know me a troll quoting stats on an unreleased box, please inform me where Apple got their stats here. I wanted to hear it from the Apple rep because I thought he gave a fairly good response to the Spec issue. I still think it somewhat misleading, but nowhere near as bad as what PC zealots were claiming.

For the record I think OSX a vastly superior OSX to either Linux or WinXP. If they improve the Finder as much as it appears I'll be a very happy camper.

But troll? Come on. Asking very reasonably questions is not trolling.

275 fps for a p4 3.0 seemed a little bit low, so I jumped over to Anandtech to see what they're used to getting. This link shows that the newest p4 (3.2) running on an nforce2 board with a radeon 9700 (not the 9800 used in the apple comparo) gets over 400 fps at the same resolution. Seems like a significant discrepancy.

I'll answer to everyone who replied to me in one big message:

G5 Specific instruction usage sounds suspicious. I really like the relaxed IEEE math operations.

These look like compiler optimisation routines, not benchmark optimisation. I agree that it DOES make a difference, but probably not such a big one.

As for the malloc special implementation, it's suspect, I admit, but I don't know enough about that to make a serious analysis, and it falls into the "of course they cheated in some places". I just don't think they cheated to the extent that's talked out in the posted article.

And yes, Altivec is in the G5, IBM incorporated the Motorola tech for them. Check here you can see the Velocity engine core.

Finally, as for the SMP bench, I tried to find them before but couldn't find my reference. Here's what I could find now for SMP and Single-CPU. The benchmark that the other guy posted are somewhat suspicious, as it shows a better increase from 2-way to 2-way with HT than from 1-way to 2 way. Not sure it makes much sense.

I'll answer to everyone who replied to me in one big message:

G5 Specific instruction usage sounds suspicious. I really like the relaxed IEEE math operations.

These look like compiler optimisation routines, not benchmark optimisation. I agree that it DOES make a difference, but probably not such a big one.

As for the malloc special implementation, it's suspect, I admit, but I don't know enough about that to make a serious analysis, and it falls into the "of course they cheated in some places". I just don't think they cheated to the extent that's talked out in the posted article.

And yes, Altivec is in the G5, IBM incorporated the Motorola tech for them. Check here you can see the Velocity engine core.

Finally, as for the SMP bench, I tried to find them before but couldn't find my reference. Here's what I could find now for SMP and Single-CPU. The benchmark that the other guy posted are somewhat suspicious, as it shows a better increase from 2-way to 2-way with HT than from 1-way to 2 way. Not sure it makes much sense.

Of course it's predictable, it's been on their roadmap for some time :)

Heck, you'll be able to "predict" the next few releases as well!

It's not as if Intel is making an 850 MHz Pentium 4.

That is, unless you count the 1.7 GHz Pentium 4 processor that safely underclocks itself by half when run fanless (or possibly by factors between half and full speed when run in a mobile setting), but even I don't count that.

There's almost no overlap (OK, there is at 1.4 GHz)

For one thing, look at low-end P4 processors vs. high-end Pentium-M processors. (Pentium-M processors, based on Intel's PIII-derived Banias core, show up in e.g. Centrino chipsets.) Intel sells Pentium-M parts rated up to 1.70 GHz, which should give a bit more overlap clock-for-clock with the P4 line, especially the 1.70 GHz Celeron 4.

What? Which part of the article was that mentioned in? Was it this part?:

It is designed to support multiple market segments and emerging applications, as a unifying I/O architecture for Desktop, Mobile, Server, Communications, Workstations and Embedded Devices. It is not just for the desktop, like the original PCI specification was designed to be.

You are completly wrong. PCI Express is completly backword compatible with PCI. RTFA.

Due to its high bandwidth, it's expected to replace AGP as well.

Actually, no. Intel (at least) will provide the X16 graphics link which will provide dedicated 8Gb/s of bandwidth exclusively for your next overpriced ATi or nVidia graphics card.

Worse than that, THG can't even get their facts straight. For example, when discussing fsutil.exe on page 4, the caption of the picture calls it a DOS app (it's not) and say it's from Sysinternals (perhaps they meant ntfsinfo, like the picture shows), yet the article text properly calls fsutil a "command line utility" (which it is) from Microsoft (which it is). While they do mention that it works on XP and not Windows 2000, they don't bother to mention that it's also available on Windows Server 2003, and that it's a system utility that's installed with the OS (c:\win[dows|nt]\system32\fsutil.exe). And just to add insult to injury, the "fsutil fsinfo" command they suggest you run is not quite correct. You need something more like "fsutil fsinfo ntfsinfo c:". "fsutil fsinfo" by itself just gives you another help screen, and not "scads of fascinating statistical information on the file system, volume and MFT."

All this article does is reinforce my dislike for Tom's Hardware Guide, and gives me ammunition I can use to convince others that THG is crap, too. If you want good hardware reviews, go somewhere good like AnandTech or Sharky Extreme. Hell, you could even go to Blue's News for the daily Hardware Reviews and still get better info. (I've not once seen Blue's link to THG from the Hardware Reviews ... I wonder why?)

The successor to the current generation of Pentium M processors, due out later this year, will have a 2 MB cache. It might be a good idea to hold off until they are out for your test :)

This is from a review of Intel's roadmap at Anandtech.

As you may or may not know, IBM originally developed Silicon-on-Insulator technology and licensed it to AMD. Here is the whitepaper: http://www-3.ibm.com/chips/bluelogic/showcase/soi/ soipaper.pdf

This is the same technology that was used to make the Power4 processor, and will also be used to make the upcoming PPC970: http://www-916.ibm.com/press/prnews.nsf/jan/06C1F2 11F9B1C24B85256ADF006163AF

AMD has recently built a new state-of-the-art fabrication facility in Dresden to produce the chips, known as "Fab 30": http://www.anandtech.com/cpu/showdoc.html?i=1773

I hope together IBM and AMD will continue to update their manufacturing process to keep on par or perhaps once again surpass Intel.

Nice trolling.

Athlon XP 2400+ beats P4-C 3.0Ghz at General Usage test

Athlon XP beats P4-C at same speed grade at Unreal Tournament 2003

Athlon XP 1700+ beats P4 3.0Ghz at RC5-72 Encryption

In the interest of fairness, Athlon XP gets beatdown by P4 at Lightwave 3D Rendering

As I said, Athlon XPs are generally better at gaming, general usage, and hard math. P4s are generally better at hard 3D rendering and media encoding.

Nice trolling.

Athlon XP 2400+ beats P4-C 3.0Ghz at General Usage test

Athlon XP beats P4-C at same speed grade at Unreal Tournament 2003

Athlon XP 1700+ beats P4 3.0Ghz at RC5-72 Encryption

In the interest of fairness, Athlon XP gets beatdown by P4 at Lightwave 3D Rendering

As I said, Athlon XPs are generally better at gaming, general usage, and hard math. P4s are generally better at hard 3D rendering and media encoding.

Nice trolling.

Athlon XP 2400+ beats P4-C 3.0Ghz at General Usage test

Athlon XP beats P4-C at same speed grade at Unreal Tournament 2003

Athlon XP 1700+ beats P4 3.0Ghz at RC5-72 Encryption

In the interest of fairness, Athlon XP gets beatdown by P4 at Lightwave 3D Rendering

As I said, Athlon XPs are generally better at gaming, general usage, and hard math. P4s are generally better at hard 3D rendering and media encoding.

From a performance standpoint, things make even less sense once we start looking at Intel's plans for the Xeon MP line for 4-way configurations. Intel's Xeon MPs will continue to use a 400MHz FSB through the first half of 2004, even as the processors reach speeds as high as 2.60GHz. Intel's reasoning behind this is mostly to allow for a single upgrade path without forcing enterprise customers to upgrade their motherboards as frequently as we have to on the desktop side, but it's clear that the performance of the processors will be limited by FSB bandwidth. Remember that Intel's MP solutions rely on a shared bus protocol, meaning that in a 4-way Xeon MP server all four processors must share the same 400MHz FSB. This essentially quadruples the FSB bandwidth requirements of the server, and the Xeon MP happens to be the processor that will have the least amount of FSB bandwidth out of all of Intel's enterprise CPUs.

Please read these :

Tom's Review of GF FX5900

Anand's Review of the same card

As you can see from the games figures, the FX5900 is faster than ATI's...

And, in the end, is the performance with applications that matters, not some stupid benchmark.

Define "convince." Anand's review shows that the FX5900 is indeed a worthy competitor for the latest ATi offering, and the superior Linux drivers put nVidia over the top for Linux users.

If you're concerned about nVidia's ethics, perhaps you should check out ATi's background.

Anantech.com has more info as does Yahoo

Yeah - that is in their 'review', which apparently needs a little fixing. Check out Hardware editor Evan Lieb's post on the forums here. They know something isn't right with their scores but they aren't yet saying what. Right now it looks as though the nVidia marks which were shown running Doom3 at high quality were actually running at medium quality. We'll see if Anandtech ever decide to publish WTF is going on with their benchmarks.

If you want better

[Next Page]

reviews that

[Next Page]

don't read like Cat in

[Next Page]

the Hat with ads, you

[Next Page]

should try

[Next Page]

AnandTech or ExtremeTech or even HardOCP.

?? I read different results elsewhere.

Check out these numbers on anandtech. Looks like the 5900 Ultra performs very well in Doom3.

~Berj

Check out this Anandtech article about the Opteron. Since none of the current Opteron MBs have AGP, Anandtech used a PCI GeForce4 and left resolution at 640x480. Further, "If you remember, up until the release of the GeForce4 Ti 4600 and the Radeon 9700 Pro, almost all of our gaming tests for CPU reviews were conducted at 640x480." So, there are people out there are doing CPU Reviews and doing what you requested.

Anandtech is still recommending a GF4-Ti4200 as an entry-level card, which can be had for around $100.

At 1024x768 medium quality no AA/Aniso, the Geforce 5900 was able to achieve 104 fps, compared to the Radeon 9800's 77 fps. Unfortunately, the game seems to be very dependent on video game performance: a Radeon 9600 could only get 40 fps and a 9200 can only get 19 fps. And remember, this is at 1024 x 768. At 1600 x 1200, the performance drops significantly. The drop is even more significant when 4X AA/8X Quality Aniso is turned on, with the Radeon 9800 only getting 42 fps and the GeForce 5900 getting 53 fps. In comparison, UT2003 gets 142 fps with the 5900 and 137 with the 9800 with the same AA settings at 1024 x 768.

Unless performance is improved drastically from the Alpha version, it looks like claims of being able to use older hardware is a little premature.

At 1024x768 medium quality no AA/Aniso, the Geforce 5900 was able to achieve 104 fps, compared to the Radeon 9800's 77 fps. Unfortunately, the game seems to be very dependent on video game performance: a Radeon 9600 could only get 40 fps and a 9200 can only get 19 fps. And remember, this is at 1024 x 768. At 1600 x 1200, the performance drops significantly. The drop is even more significant when 4X AA/8X Quality Aniso is turned on, with the Radeon 9800 only getting 42 fps and the GeForce 5900 getting 53 fps. In comparison, UT2003 gets 142 fps with the 5900 and 137 with the 9800 with the same AA settings at 1024 x 768.

Unless performance is improved drastically from the Alpha version, it looks like claims of being able to use older hardware is a little premature.

At 1024x768 medium quality no AA/Aniso, the Geforce 5900 was able to achieve 104 fps, compared to the Radeon 9800's 77 fps. Unfortunately, the game seems to be very dependent on video game performance: a Radeon 9600 could only get 40 fps and a 9200 can only get 19 fps. And remember, this is at 1024 x 768. At 1600 x 1200, the performance drops significantly. The drop is even more significant when 4X AA/8X Quality Aniso is turned on, with the Radeon 9800 only getting 42 fps and the GeForce 5900 getting 53 fps. In comparison, UT2003 gets 142 fps with the 5900 and 137 with the 9800 with the same AA settings at 1024 x 768.

Unless performance is improved drastically from the Alpha version, it looks like claims of being able to use older hardware is a little premature.

Read this article NVIDIA's Back with NV35 - GeForceFX 5900 Ultra

3Dmark03 may be inflated but what counts is real world game benching. And FX 5900 wins over ATI in all but Comanche 4.

Interesting ehh?

Check Anandtech review here.

Currently mid-range Geforce FX 5600 Ultra get 50fps at 1024x768 resolution. Final version will run faster.

Anandtech's Doom3 benchmarks show Nvidia has the crown in Doom3 by a fair margin. And probably will for most pixel/vertex shader engines. If you were going to spend $500 on a card I don't know why you'd get the ATI 9800 Pro.

Here.

Geforce FX show some impressive performance. The good news is that even mid-range GeforceFX 5600 Ultra gets respectable framerates (55fps) at 1024x768 resolution.

In the extremetech review, the version of the drivers used were not specified.
What kind of review is that?
ohh, wait, is this some of those comercial /. posts?
ahh, now I understand why anandtech.com or tomshardware.com links were not in the original post. ;-)

Click and learn:
Tomshardware review
Anandtech Review
Nvnews review/news
Review links

In the extremetech review, the version of the drivers used were not specified.
What kind of review is that?
ohh, wait, is this some of those comercial /. posts?
ahh, now I understand why anandtech.com or tomshardware.com links were not in the original post. ;-)

Click and learn:
Tomshardware review
Anandtech Review
Nvnews review/news
Review links

Anandtech and Tom's Hardware have much better hardware reviews than that ZD reviews specified. They also have Doom 3 bench marks, which put the new NVidia card significantly ahead of the ATI counterpart.

Anandtech and Tom's Hardware have much better hardware reviews than that ZD reviews specified. They also have Doom 3 bench marks, which put the new NVidia card significantly ahead of the ATI counterpart.

Allow me to speculate.

Barton (Model 10) comes in three flavors: 3000+ (2.167 GHz), 2800+ (2.083 GHz) and 2500+ (1.83 GHz). All other things equal, the 3200+ should run at 2.25 GHz, same as the 2800+ Thoroughbred (Model 8).

However, if AMD were to increase the FSB speed, you can expect the CPU frequency to be slightly lower. I would guess between 2.083 and 2.167 Ghz.

AMD keeps a definitive list up to date.

Hey, Guspaz the anus licking fag; try using "A HREF"

http://anandtech.com/cpu/showdoc.html?i=1818&p =6

See, pussy? And the extra space in the URL doesnt affect the link target, just they way it looks. FuckCode SlashCode does that.

Lick sack, homo.

You're correct, I've read the same article. It was from the Anandtech tour of an Nvidia plant I'm 99% sure. They looked at a pre-fab lab and had a large IKOS box there that they were emulating the NV30 core on at about 1/10000 or lower speed ratio. Article is located here on Anandtech.

Basically what it boils down to is it would work great for lower performance applicatons like sound and such, but would fail miserably at any higher demand applications such as video processor.

Anandtech also mentioned it but says For the record, we did not encounter any issues during our testing that were out of the ordinary. Our motherboard labs did encounter some issues early on and we'll be investigating to see if they could possibly be related, but without being told any symptoms or the nature of the issue it's going to be very difficult for us to figure out what's causing it. Luckily none of these CPUs should have made it into the hands of any end-users, which is a relief. Intel insists that the problem is limited in nature, only time will tell how much that holds true

But I haven't heard anything specific (as of yet) as to WHY they are pulling them back. At least Intel is making sure that there's a quality product on the market.

It depends on the monitor and your own tastes. If you look around, you can find monitors that are ok on response time--definitely better than typical laptop displays. The monitor I mentioned has a 25ms response time, and there are already monitors that beat it. In practical terms, you can definitely tell the difference between this monitor and a good CRT, if you're looking for it. But I think my brain interprets the blur as a slight "motion blur", because when I'm in the midst of a game, I never notice it. :)

Additionally, the monitor is nice in a lot of other ways that I wouldn't want to give up for a CRT. Of course, you get what you pay for. Viewing angle, contrast, brightness, and response time vary wildly from the low end to the high end, so you have to shop around.

Anantech reviewed a Hitatchi monitor that advertises 16ms response time, which should be quite excellent. Unfortunately, it sounds like the reviewer still noticed some blurring. The monitor sounds pretty nice in spite of that, though.

If you're thinking that LCD monitors won't be common at lan parties, it appears not to be the case. People have all sorts of equipment and computer accessories that you can peruse, some of it of a rather dubious quality. As it looks "cool", it gets purchased.