Slashdot Mirror

Click Here

I suppose that depends on your tests. How much attention should we pay to the databse test if the review says:
We expect to see a performance increase with HyperThreading - SQL servers must thread well.

Yet the sql-bench manual says:
Note that this benchmark is single-threaded .

Take a decent benchmark, run it with 10 instances simultaneously and then we'll see if it really helps for heavy multithreaded loads. (That doesn't really belong in a desktop test, but does compiling?)

Printer friendly link

Tired of clicking next page

From the first line in TFA:

A few weeks ago, we introduced the first of a series of articles on building a home made PVR, "Building a Linux PVR Part I - MythTV Setup and Install"

They "glossed over it" because this is the second article; the first one is about the actual setup and installation for MythTV.

Even better - any of the Prescott P4s!

Well, here you go... AMD's fastest Opteron runs neck and neck against Intel's fastest Xeon, and guess which one is cheaper? (Only slightly though...)

*(i wouldn't personally count office benchmarks like word but i know intel has a weird history of doing well in those)

Since the implementation is essentially two cores on one die, the speed would be limited by whats available with the existing 90nm line at present. If you were asking about their performance rating, I'm guessing it might be way higher than the existing line of Opterons.

..remember, this may be a Server processor, but it is also the close cousin of the PC dual core offering, see http://www.anandtech.com/cpuchipsets/showdoc.aspx? i=2178&p=2 . So, we will be able to see if it is worthwhile to spend bucketloads of money on a 939 socket athlon motherboard today.

This was only true for a very short period of time on paper, and is no longer true on the consumer market. The first IA-32e/EM64T enabled Pentium 4 Intel is shipping is the P4 "Prescott" Model F, which is based off of the E-0 core. The E-0 core not only supports EM64T(which was added in the D-0 core for Xeons), but eXecute Disable(XD) support, Intel's version of NX, was added in the latest core(AnandTech has the full details). We should see the first P4F's this September.

Anandtech originally posted an article which was a comparison of the Intel Xeon Nocona ("3.6F") and the AMD Athlon 64 3500+. The Xeon "won" most of the benchmarks by a good amount.

The criticisms were that the Xeon is not a desktop CPU, or vice-versa, the Athlon is not a workstation/server CPU. But are they so different? The Xeon has 1MB L2 cache, and so does the P4 Prescott (and presumably Prescott with x86-64 enabled), and both run at the same speed.

Similarly, the 3500+ runs at 2.2GHz and has a 512KB L2 cache, whilst the Opteron 250 runs at 2.4GHz and has a 1MB L2 cache.

With this in mind, can anyone explain to me why the Opteron seems to perform much better? The benchmarks appear to show the Xeon trouncing the A64 and the Opteron 250 trouncing the very same Xeon in 64-bit.

So what's the deal? Are AMD's desktop chips choked on 512KB of L2 cache (and yet the new Socket 939 A64s seem to be dropping back to 512KB L2 cache, whereas a few of the Socket 754 chips had a full 1MB).

I'm no processor expert, I just wondered if anyone could explain the "big difference".

No message here. Oh, did you know that an Athlon64 3000+ is within 2fps of a P4 3.4 Extreme Edition in Doom 3?

Look up the prices for those two items.

Given the fact that most people have only AGP ports, this is a barrier to adoption. It has been reported that AGP versions will follow.

So to produce an AGP version of a PCIe-native GPU like the GeForce 6600, all NVIDIA needs to do (theoretically) is turn the HSI the other way. Since NVIDIA has stated that AGP versions will follow "shortly afterwards" (according to this Gamespot article), I don't think the PCIe/AGP issue will be a barrier to adoption.

Here's a nice, simple Anandtech article on HSI: NV45 Preview: On Package HSI

Given the fact that most people have only AGP ports, this is a barrier to adoption. It has been reported that AGP versions will follow.

So to produce an AGP version of a PCIe-native GPU like the GeForce 6600, all NVIDIA needs to do (theoretically) is turn the HSI the other way. Since NVIDIA has stated that AGP versions will follow "shortly afterwards" (according to this Gamespot article), I don't think the PCIe/AGP issue will be a barrier to adoption.

Here's a nice, simple Anandtech article on HSI: NV45 Preview: On Package HSI

Like this you mean ?:
Intel and AMD

Like this you mean ?:
Intel and AMD

Taken from here

17 - Posted on Aug 9, 2004 at 5:32 AM by KristopherKubicki
The only reason we even put the 3500+ in there is cause we already had benchmarks for it.

Relax, its just a primer for future articles. A 3.6F is supposed to compare with a "3600+" rated Athlon 64 isnt it? Since we dont have a 3600+ the 3500+ should perform slightly lower? Isnt this what we expected? And for those of you who dont believe me, a 3.6GHz 1MB EM64T Nocona is *exactly* like a 3.6F.

I thougth the AMD chip did pretty damn good for costing $500 less!

Kristopher

So should I save up for an Intel processor or buy 2 AMD machines?

The XPS is old AND it has been reviewed by better sites. Review

What planet are you from to believe that a 3GHz P4 "stomps all over(perhaps all but the very newest) AMD offerings" ??

Have you read any benchmarking articles inthe last year or so?

Doom 3 CPU Benchmarks
THE Graph

Even the low-budget "Sempron" chips beats a 3.0GHz P4. Not even the 3.4GHz EE P4 can stand up to a 2.2GHz Athlon64 3500. Or was the phrase "all but the newest" supposed to rule out AMD chips introduced in the last 12 months? Spread your FUD elsewhere..

What planet are you from to believe that a 3GHz P4 "stomps all over(perhaps all but the very newest) AMD offerings" ??

Have you read any benchmarking articles inthe last year or so?

Doom 3 CPU Benchmarks
THE Graph

Even the low-budget "Sempron" chips beats a 3.0GHz P4. Not even the 3.4GHz EE P4 can stand up to a 2.2GHz Athlon64 3500. Or was the phrase "all but the newest" supposed to rule out AMD chips introduced in the last 12 months? Spread your FUD elsewhere..

Also, Anandtech just posted a new roadmap with some info on upcoming 64-bit Pentium 4 CPU/chipsets for the desktop. The Intel 925XE chipset (with 1066MHz FSB) will ship in October along with 64-bit Pentium 4 "F" processors. "F" supposedly means it's a 64-bit Prescott.

Though Xeon-Nocona has been available for more than a month it seems there there are no substantial reports on 64-bit performance of Nocona. Is there anyone here who can report anything about the 64-bit performance of Nocona?

Those screenshots are the relatively useless SiSoft Sandra 2004 Memory scores. In addition, the Athlon 64 3000+ is only using a single channel of DDR memory, compaired to the dual channel P4 system. If you actually look at benchmarks of the Athlon 64 3000+, you will see how well it performs.

Here is a review that has more useful benchmarks to compare P4's and A64's.

Don't be so certain about that. Check out this and this. Intel is having production problems. They had to do a recall on the i925, availability of the 3.6ghz P4 is so low that even their bread-and-butter OEM Dell is having problems supplying the chip for their top-tier systems(to the point that the 3.6 ghz P4 is currently, effectively, a paper release), and they have bugs in the Lindenhurst chipset that will likely put a damper on anyone wanting to adopt this technology(other than Dell). I suspect that, if their 3.6 ghz desktop P4 is not coming out of production in large numbers, they may have similar problems producing Nocona CPUs as well.

Don't be so certain about that. Check out this and this. Intel is having production problems. They had to do a recall on the i925, availability of the 3.6ghz P4 is so low that even their bread-and-butter OEM Dell is having problems supplying the chip for their top-tier systems(to the point that the 3.6 ghz P4 is currently, effectively, a paper release), and they have bugs in the Lindenhurst chipset that will likely put a damper on anyone wanting to adopt this technology(other than Dell). I suspect that, if their 3.6 ghz desktop P4 is not coming out of production in large numbers, they may have similar problems producing Nocona CPUs as well.

Don't be so certain about that. Check out this and this. Intel is having production problems. They had to do a recall on the i925, availability of the 3.6ghz P4 is so low that even their bread-and-butter OEM Dell is having problems supplying the chip for their top-tier systems(to the point that the 3.6 ghz P4 is currently, effectively, a paper release), and they have bugs in the Lindenhurst chipset that will likely put a damper on anyone wanting to adopt this technology(other than Dell). I suspect that, if their 3.6 ghz desktop P4 is not coming out of production in large numbers, they may have similar problems producing Nocona CPUs as well.

Really, I'm not sure why they are bothering with XP-64. Longhorn is due out soon enough... I'm just not sure I'm interested in paying for the product that will come out just before Longhorn. It's like if you had a choice between buying a flintlock pistol or a single-shot bullet operated colt, when you could wait and spend a little more money on a colt six-shooter. My point is that there's not much difference between XP and XP-64 compared to XP and Longhorn. I'm moderately satisfied with XP, apart from all the annoying Microsoft crap that comes with it, and there's no telling how much *more* of that will ship with XP-64 or even Longhorn. So I wouldn't be upgrading to get rid of the annoyances in Microsoft's products, just in some hopes of better features! I wouldn't hope for better security in future Microsoft products, because that would be futile, IMHO. The best solution for going 64 today looks like a Linux!

Of course Nvidia's card is going to do better. Doom3 has a specialized codepath for nvidia hardware, while the ATI card does not.

Ofcourse the non-nvidia render path is optimized for matrox, xgi, s3 and bitboys cards... I hear carmack has been spending a lot of time optimizing for intel chipsets with integrated video ;-)

Seriously though, ATI has never done openGL really just *right*. For years you could not play half-life on their cards using openGL... (glquake dll worked with a hack but this didn`t solve everything) They seam to be having an edge on DirectX 9 games though.

However, these benchmarks are not that discouraging for ATI as you might think. They show that no mather what contemporary card, you will get a playable doom III game. This is great couse many people will have a loyalty crisis coming, not in ATI vs nvidia or DX9 vs opengl but in building the either the ultimate doom3 or half-life 2 machine. It seems making a decent half-life machine is gonna be a lot more difficult and doom is gona be a lot more forgiving. Its nice that doom doesn`t require a diffrend card to be swapped in before playing no mather what halflife machine you get. Also I think half-life 2 will be a better benchmark to base you decisions on becouse very few games will have an engine like doom III in the coming years. I am really looking forward to games on the doom III engine though. It would be great if another valve stood up and did a somewhat realistic tactical singleplayer optimized FPS with storyline and good storytelling in it. I mean environments that make sense, death to cutscenes! Upside down flaming skulls arent really my thing...

I would suggest checking out this thread

From what I can tell, most heatsinks will not work without some modification (sometimes as simple as adding washers), but read the thread, it answers a lot of questions.

I hope I'm not the only one who caught this, but it appears the MEncoder version number used in the tests, 3.3.1 doesn't exist.

Anandtech page: http://anandtech.com/linux/showdoc.aspx?i=2114&p=3
I've used mencoder a lot with Linux, and I know the latest version, other than the CVS, is 1.0pre4. When the version is custom compiled, it appends your gcc-version number to the end of the version, so it could become 1.0pre4-3.3.1.

MPlayer/MEncoder page: http://www.mplayerhq.hu/homepage/design7/news.html
Curiously, the next page of the Anandtech article shows rendering tests with Mental Ray version 3.3.1.

Second Anandtech page:http://anandtech.com/linux/showdoc.aspx?i=211 4&p=4
Interesting! The very same version specified incorrectly for MEncoder! I think we may be dealing with a minor mistake in this article, though frankly, I can't imagine getting those kinds of speeds from MEncoder with decent quality and bitrate. My AMD Athlon XP 2100+ only gets 13 FPS, and I know that mencoder doesn't take advantage of the x86-64 instruction set.
Oh well, just thought I'd bring this to /.'s attention.

I hope I'm not the only one who caught this, but it appears the MEncoder version number used in the tests, 3.3.1 doesn't exist.

Anandtech page: http://anandtech.com/linux/showdoc.aspx?i=2114&p=3
I've used mencoder a lot with Linux, and I know the latest version, other than the CVS, is 1.0pre4. When the version is custom compiled, it appends your gcc-version number to the end of the version, so it could become 1.0pre4-3.3.1.

MPlayer/MEncoder page: http://www.mplayerhq.hu/homepage/design7/news.html
Curiously, the next page of the Anandtech article shows rendering tests with Mental Ray version 3.3.1.

Second Anandtech page:http://anandtech.com/linux/showdoc.aspx?i=211 4&p=4
Interesting! The very same version specified incorrectly for MEncoder! I think we may be dealing with a minor mistake in this article, though frankly, I can't imagine getting those kinds of speeds from MEncoder with decent quality and bitrate. My AMD Athlon XP 2100+ only gets 13 FPS, and I know that mencoder doesn't take advantage of the x86-64 instruction set.
Oh well, just thought I'd bring this to /.'s attention.

I would suggest that you have not read the article yourself, and have merely skipped to the conclusion (which is rather an odd one, seeing as it does not quite reflect the benchmarks - they actually split both the gaming and rendering).

Take a look at, for example, this benchmark, where Windows outright beats Fedora at both 32- and 64-bit, and only loses to 64-bit SuSE slightly because it doesn't have a 64-bit binary itself, and this one, where Windows just plain wins.

I did mess up on the "Windows wins at rendering" part, though, sorry for that - they split it actually. I didn't notice the "lower is better" part on the Mental Ray bench and just went with the one that had longer bars. Oops.

I would suggest that you have not read the article yourself, and have merely skipped to the conclusion (which is rather an odd one, seeing as it does not quite reflect the benchmarks - they actually split both the gaming and rendering).

Take a look at, for example, this benchmark, where Windows outright beats Fedora at both 32- and 64-bit, and only loses to 64-bit SuSE slightly because it doesn't have a 64-bit binary itself, and this one, where Windows just plain wins.

I did mess up on the "Windows wins at rendering" part, though, sorry for that - they split it actually. I didn't notice the "lower is better" part on the Mental Ray bench and just went with the one that had longer bars. Oops.

There's a huge market segment that purchases equipment that's one step below bleeding edge just for the price break. Could it be that there's more money to be made by soft locking a single product than to manufacture multiple products?

This sorta reminds me of the boon brought by the Celeron 300A

When I put together my last machine in 2000, I decided to break from the path of buying whatever shit case presented itself under 20 bucks, and spend a good chunk of change on a
It is the best piece of computer gear that I have ever bought. I expect to use this case for as long as the ATX specification is standard in mainboards. That should be your main criterion for purchasing a case- spend a good deal of money on it, because if you're posting on slashdot, you are probably enough of an enthusiast that any PC case you own will house several different boards and processors over its lifetime. This one is certain to stand up to many upgrades, and it won't annoy me by forcing my fingers into contortions or slicing them on sharp edges while I'm fixing it.

Side 1 - people suck - have you ever read any of these "deal forums" (Fat Wallet and AT Hot Deals are two I read from time to time. Jesus fucking christ these people would walk a mile through the desert barefoot while eating salted peanuts for a fucking $10 rebate. Sometimes the "sequence of events" required to get these deals is more fucking work than just well, working and BUYING it for regular price, pricematch here, rebate there, obscure coupon code over there, print someone's reciept from some other store in some other state, and you might get a $79 item for $64 or something equally lame. I get a chuckle at the enormous lengths people will go to to save a couple bucks.

Side 2 - Best Buy sucks - reading this site is like a traffic accident complete with ripped off limbs, you can't stop looking but you know you should. The damn near criminal "support policies" they push and push and push on you, and the "piggyback" magazine subscriptions make me want to do things that would get me on the news. I'm currently getting bills from Entertainment Weekly because when checking out at BB a few months ago I let them "send me 4 free issues" just to get them to shut the FUCK up. Now they want me to pay for the stupid magazine like I give two drops of spit about EW (ew is right).

I don't know who to root for in this fight... Best Buy or the "demon customers" socking it to them. Goddamn I hate rebates, I go out of my way to avoid deals involving rebates because they just PISS ME OFF so bad.

Really I'm not ranting...

If by reliable components you mean reliable powersupplies, there are a few brands which are well known to be high quality and reliable.

Antec is considered to be the top end for reliability and performance. They contain seperate transformers for the different voltage rails. I have 3 Antec powersupplies in my computers. All have worked great.

Enermax is another maker of very beefy powersupplies. I've got one and haven't had a problem with it.

There's bad news, though. 50% premium? No. Try 200%, if you're used to those shitty $30 powersupplies. A 380W Antec will set you back somewhere in the region of $90. It's worth it, though. Cooler powersupply, cooler system, increased stability due to lower temperature and solid voltage.

Some reviews at Tech-report and AnandTech should give you some baselines to look at.

In fact, we installed 10 workstations using this system on 5 PCs for a client of ours recently--replacing 10 old iMacs--to lower the TCO for a small call center.

It's been working great, no problems whatsoever.

Well, you can, but it's third party, very hardware specific and leaves you stuck with M$ XP. The Linux system demonstrated is a clear winner for schools, libraries, banks, casinos and other places where economical use of hardware is desired. While the Linux system might be difficult to maintain, it can be done. The Windoze solution leaves you dependent on the vendor. The people at Jetway have done an outstanding job but such is the world of proprietary software.

Anyone know of a better system? If you say citrix terminal services, I say, Linux Terminal Project and, cool as those things are, get thee back to 1975.

The fastest $117 2.8ghz celeron got the shit kicked out of it by a lowly $55 Athlon 2400XP. Who in their right mind would buy one of these chips?

Here's how the Celeron D 335 actually performed against the XP 2400+:

What I'm really waiting for is Celeron D on LGA775 and Sempron on Socket 939.

AnandTech has a more thorough review

You must have a magic Parhelia. 70fps is plenty good (assuming it doesn't dip much below 50-60, which I doubt), but frankly I don't believe any Parhelia has ever done 70fps at 2400x600 in UT04, unless maybe you minimize every detail option, stand still, and stare at the ground or a wall.

2400x600 is 1.44Mpixels. That's more than the 1.31Mpixels at 1280x1024, under which conditions the parhelia cant break 40fps in UT03 according to Anandtech's benchmarks.

I can't find any Ut04 benchmarks for the Parhelia (probably because it's been so long since any gamer really cared about a Matrox product), but I can't imagine it gets any better, unless again, of course, you minimize everything and don't move.

When we can find a 1.6GHz Duron for just over half the price of a 2.6GHz Celeron and get better performance consistently in almost every test we ran, the choice is clear.

When we can find a 1.6GHz Duron for just over half the price of a 2.6GHz Celeron and get better performance consistently in almost every test we ran, the choice is clear.

When a single 6800 card requires a 480watt power supply and two dedicated power lines, what would the power requirements be for two of these cards in the same computer system?

There are very little volatile-memory related software bugs.....

Oh, are you SURE about that? You should research such statements first, my friend, rather than assuming.

Take a look at this review from last year of power supplies by Anandtech.

They ran a six-hour memory test 54 times--and found that with 512MB of RAM, after each six hour test there were an average of four bits that had flipped! That means there is a memory error on a 512MB PC--on average--every 90 minutes!

If that error occurs in a code segment in a driver, you may get a system crash. In a Windows DLL, perhaps some system instability. In an application, perhaps an application crash. If it's in a data segment, your important manuscript may suddenly lose a paragraph or skip a couple pages as a linked list pointer jumps to the wrong spot, or you may find a bunch of junk replacing normal text.

Memory errors are a serious problem that very few people acknowledge. Why people still buy non-ECC RAM is beyond me. (Of course, even with ECC RAM, there are still various places inside the PC where failure can occur--along the various buses for exmaple, which don't all have ECC. So this is only part of the solution.)

More reliable RAM would definitely be a step in the right direction.

In the US, T-Mobile does unlimited WAP GPRS for about $5/mo (in addition to your normal talk plan.) There's also unlimited unfettered GPRS (eg no port blocking) for $20 per month, or $30 without a talk plan, if you plan to use it with your laptop.

Actually, you should be able to get that GPRS Internet access at no extra cost. It's not something they advertise, but if you set up a connection through the phone that dials *99***1# (or maybe just *99#) and logs in with no username and no password, you'll have full 'net access at dialup-equivalent speeds. I use it to browse websites and log into my home server (over SSH) to check mail and news.

Anandtech's Review of the HDTV wonder; Anandtech has been around a while, and is still one of my hardware-review sites of choice.

So does Anandtech.
Anandtech review