Sign in with
Sign up | Sign in
Your question

Upgrade now or wait for HammerTime?

Last response: in CPUs
Share
September 5, 2002 12:19:57 PM

Hi,

My current system is getting a bit long in the tooth, so I'm thinking about upgrading to a new system. Should I upgrade to an XP 2600/2800, or is it better to wait until the ClawHammer is released?

BTW does anyone have any news on when the CH is due for general release?

Thanks,

Jim
September 5, 2002 2:05:33 PM

Personally I never buy what's brand new, it's the price/performance ratio I don't like. And as for waiting, there will always be something better around the corner. And we don't know (or do we) just how well the AMD 64 bit CPUs are going to work for us in regular applications.

_______________________________________________________________________
TeeCee
<A HREF="http://www.digitalfreaks.net/" target="_new">Digital Freaks </A>
September 5, 2002 2:07:31 PM

First, you won't find any XP2600 nor XP2800 now. XP2600 can be some weeks from now, but XP2800 it will take more time.

CH it's supposed to be released in 1Q 2003. Well, if you can still wait to 2003, you will have the next plataform and a clear upgrade path. Anyway, a XP2600 will last you for 2 years or still upgrade to Barton.

DIY: read, buy, test, learn, reward yourself!
Related resources
September 5, 2002 2:19:17 PM

Well Jim, if you're upgrading soon you might want to consider getting a 2800+ or better. The 2800+ will be using the Athlon XP Barton core, which offers a faster FSB and more cache. While AMD says they'll be out in October, I frankly don't believe them. Case in point; where in the hell are the 2400+/2600+? So maybe around thanksgiving. This processor, when paired with a decent board should perform far times better than current AXP's since it offers more cache and other optimizations.

Clawhammers *may* be out by christmas. If they are, the supply will most likely be low, with a huge demand. So they will be next to impossible to get without a large bank until there's plenty to go around. Also the new boards for the CH will just be debuting, with a price premium to match. Also we don't know how the CH is going to perform yet. The 2800+ barton's will use the same socket A current athlons use. Any recent board said to support "2200+ and future chips" should run the barton axp fine with a simple BIOS flash.

You might also want to consider intel. Intel processors, boards and even rambus are almost at par with high-end athlon configs. I recently configured both a high-end athlon and very high-end pentium 4 system, and the undoubtly better performing p4 system only was $150 more.
September 5, 2002 3:22:42 PM

Thanks for the good advice - I think I'll wait until the first Barton core AMDs are released and see how they measure up, and just how much they cost.

Jim
September 5, 2002 4:09:17 PM

Buy a P4 now or wait 1 year

At the end i have speak with a horny lady
September 5, 2002 5:39:41 PM

What are you going to be using the computer for? Do you use any software that would take advantage of the newest PC about to be released? Is it worth that wait? What's wrong with what we have out now? Does it not meet your needs for the next couple of years?

Unless there's some specific reason why you'd want to wait another 6 months then I'd go for the upgrade now. You'll always be waiting otherwise. There's always something cool around the corner.

<font color=red>I'd like to dedicate this post to all my friends, family, and fans. Without them this post would never have been possible. Thank you!</font color=red>
September 5, 2002 5:51:20 PM

Buy a P4 now or wait a year? Are you insane? I'm trying to be nice here, but what kind of p.o.s. advice is that? First off, if you buy now, buy the 2000+ if you like the price/performance thing. That's if you can't wait until the 2600+ hits the market (WHERE ARE YOU?!?!). If you can wait for it, I agree that the 2800+ will rock. The 2700+ will have the 333MHz FSB, and the 2800+ will step up to a 512k L2 cache, which by the way AMD should have put in a long time ago. As for the P4, RDRAM is close to the same price... PER STICK. But remember, you have to have at least two sticks to run dual channel RDRAM, otherwise you're better off running DDR SDRAM.

Currently, Intel wouldn't get my vote just because they still are lightyears behind AMD in the price/performance category. The only things Intel have going for it are:
1) the most stable chipsets in the world, and
2) the most overclocking headroom.

Neither of this matters if:
1) you have enough computer knowledge (VIA's bad, but not that bad)
2) you aren't going to overclock

I don't overclock, and I don't have a ton of money to waste. Every penny counts. And, in fact, we're not talking about pennies. I've already gone over this at the Anandtech forums, so I'll just repost the numbers. The systems I priced were based on THG's setups for the AMD and P4 systems in its 2.8 P4 review. All prices were from Pricewatch as of two days ago:
"The P4 setup would cost $927 for the 2.8 P4 setup and $630 for the 2.533 P4 setup, both including shipping. Mind you, that's using Samsung instead of the Kingston that they used, which would have been $46 more with no difference in performance. I am fair. On the other hand, the Athlon 2600+ (which is preordering for right now for $300) setup would cost $494, while the 2400+ setup (again it's on preorder) would be $394. That is using Samsung RAM as well, since I couldn't find ANY winbond in 512MB sticks. You were asking where the $500 - $600 differences were? Well there you have it. Just in case you want to see like-performing systems, the same systems with a P4 2.2Ghz and an AXP 2200+ would cost $587 and $339 respectively. Nearly a $250 difference! I don't care how much money you have, you're crazy to spend an extra $250 just to have the P4 name. There's not enough overclockability or stability to justify that in my mind."

Well said if I do say so myself. I'm considering upgrading a little now (ECSS K7S5A which is a SiS 735 chipset mobo, and an Athlon 1600+) as an intermediary to the Hammer, which AMD is swearing will be out by Christmas. However, I would also agree that it's best not to buy as soon as it comes out. Not only will you pay a HUGE premium, but you won't have a bunch of reviews to look at to determine which HSF is best, or which chipset/mobo is best. It's all up to your discretion, but as somebody said earlier, there's always something better around the corner no matter when you buy.

She said "I love a man in tight jeans" and I said "They're not supposed to be tight I just got fat."
September 5, 2002 5:52:40 PM

My PC is mainly used as a high end Games platform, and I want to buy a system that will last a couple of years without much upgrading - my main worry is if I buy a system now, how big a performance jump will the ClawHammer bring, and will it prove more cost effective to wait for it's release?

Thanks,

Jim
September 5, 2002 5:59:59 PM

On the higher end of the CPU scale (2.4B and beyond), the P4's price/performance ratio is actualy not that bad compaired to AMD's, and there are plenty of other nice chipsets out for the P4. AMD's value is at the lower end of the spectrum (1800+ and lower). Faster than that, and you can get a nice P4 B for a similar price and get similar performance stock.

AMD needs to get in gear and make some actual improvements to the core before they find themselves left in the dust.

BTW, I keep on hearing that Barton will have a 166 FSB. Where is info on that? Or is it just speculation?

If ignorance is bliss, then why is everyone so miserable?
September 5, 2002 6:01:46 PM

It might prove more cost effective to wait till it is released and snatch a good AMD/Intell CPU when they drop in price. When the new chip is released, the previous best usually has it's largest price drop, and it's not a bad time to buy.

If ignorance is bliss, then why is everyone so miserable?
September 5, 2002 6:18:57 PM

A year, 6 months? WHAT??? 3 gig pentiums w/ hyperthreading will be around for x-mas, maybe sooner knowing intel. Also, the barton XP and the nforce2 will prolly be AMD's best offering at this time. Okay let's check this out. Let's skip ahead a month and imagine AMD finally gets the 2600+ out. Maybe even the nforce2 board.


ATHLON BARTON XP 2600+ approx: $360?
COOLER: VOLCANO 7+ $25
BOARD: WITH NFORCE2 CHIPSET n/graph or something else very high end from performance approx: $150
RAM: 512MB Kingston CL2 approx: $150
HARD DRIVE: WD800JB 80GB/8mb 7200 $105
PSU: Enermax EG365P-VE (350w) $65
DEVICE 1: FAST DVD $40
DEVICE 2: 32X CDRW $50
SYS: $950 w/o sound, video and monitor since I have those.

---Or---
CPU: Pentium 4 / 2.66GHz $440
BOARD: Asus P4T533C $160
RAM: 2/256MB Samsung PC1066 RDRAM $240
HARD DRIVE: 3WD800JB 80GB/8mb 7200 $105
PSU: Enermax EG365P-VE (350w) $65
DEVICE 1: FAST DVD $40
DEVICE 2: 32X CDRW $50
SYS: $1100 w/o sound, video and monitor since I have those.

I'd totally take the P4 with the RDRAM for only a $150 more. Plus the pentium system prices are current! So most likely they'll be even more on par by the time we see a 2600+. So where is this performance/price rational coming from ppl?
September 5, 2002 6:38:27 PM

Price performance only exists at 2200+ and down, where AMD performs on par or better than Intel 2.2ghz and lower procs. Recently the people here who support AMD have said "you dont need all that power right now" when referring to anything above 2200+. That's a true statement that applies to the avg user. But it's really a stretch to find a reason to get someone to buy AMD procs.

This sig runs too hot.
September 5, 2002 7:27:05 PM

Well you prolly won't need all the power the clawhammers offer either, but I'm sure they'll get one as soon as they can. I currently have an AXP 1900+ system, but the way it's looking to me is an extra $150 for stability of the chipset and performance is totally worth it. It used to be $500 or more difference. Even if I were to change the above configuration with lower offerings, such as a 2200+ ($150) and 2.26GHz ($197) the price difference is under $50. I'm just saying AMD's price/performance sweet spot is much harder to come by anymore. And as I illustrated above, even the current, high-end, p4 systems are only dollars away from AMD's next round of offerings, and by the time they're out the prices of the current intel set will likely fall even more bringing it almost to par price wise.
September 5, 2002 7:59:11 PM

this just in...."Intel proudly announces its new line of value processors....the Pentirons" May be reality in less time than you think.

Just because you're not paranoid, doesn't mean they're not watching you.
September 5, 2002 8:03:07 PM

As LED and nja469 have already stated, the price/perfromance only exists at XP2200+, and lower. Athlon's are no longer the high-end CPU's of choice. I suggest you stop spouting BS and stop being such an AMD fan. Even the XP2200+, now has some fierce competition from the 2.2B & 2.4B P4's. The XP2200+ is $145 right now on pricewatch, while the 2.2B and 2.4B are $196 & $197, respectively. So, obviously, there's a point to getting the 2.4B because it costs the same as the 2.2B. The 2.4B, also offers more performance the the XP2200+ (not alot). And the price difference is only $50, so even for AMD fans, it's become a tough decision as to which CPU to get.

And let my add something else the Intel has going for itself:

- the most stable CPU's (more stable than AMD's)
- alot of safety measures on the CPU's, thus giving more stability to the CPU

Quote:
<i>originally written by Kelemvor</i>
Neither of this matters if:
1) you have enough computer knowledge (VIA's bad, but not that bad)

That is such a stupid comment. That means if you have enough computer knowledge then you <b>don't</b> need a stable chipset? That's total BS. WHY wouldn't you want a stable chipset?

jamessiddle, as several people have already mentioned, no matter when you buy, there will always be something around the corner. If you really want to, you could wait till next year and get either a Clawhammer, or a P4 Prescott. The Prescott will be a 3rd P4 revision, while clawhammer will be a next generation CPU. If you get either a clawhammer or Prescott, you may get a better upgrade path for future upgrades.

- - -
<font color=green>All good things must come to an end … so they can be replaced by better things! :wink: </font color=green>
September 5, 2002 8:23:11 PM

Quote:
And let my add something else the Intel has going for itself:

- the most stable CPU's (more stable than AMD's)
- alot of safety measures on the CPU's, thus giving more stability to the CPU

Stop spouting FUD, Dark.
Most stable my ass, even if Intel does rigorous testing, while AMD does not as much, please link me to a recent post or world known problem where an AMD proc. started crashing, doing bad mathematical problems. Oh wait, Intel did that! Really, that is just fanboy-expected comments, which you shouldn't have added, because it's FUD and in fact is ironic. In fact, a company called Samson or something like that, started using 100 AthlonMPs in clusters, another company also did similar things for scientific or space stuff. I wonder why Intel wasn't used? Not because Intel has previous problems but because they don't HAVE to choose Intel each time when looking for that so-called ironic stability.

Lots of safety measures, yeah that's a good thing, but AMDs also have thermal protection with the new diodes, and have been tested by an AMD video, where they played for 9 minutes long under a no fan but HS on, until it turned off, and when they took off the HSF completly, the status LED blinked and turned the comp off, proving that it works. Only ASUS COP was reported to have probs with booting without a fan, apparently, but since it was by Asus' own design not AMD's, PLUS the fact only an ultra-idiotic moron would run a new PC without a fan, I think it's safe to say that buying an Intel solely because it has oh-so-toutable safety when AMDs also do that, and neither would last longer than the other (unless the silicon itself was low quality), is, as Matisaro would say, RETARDED.

Chances are 1 in 1000 or higher, and even if it happens, either processors, AMD using the newer mobos since June 10th or the A7V333, Soltek 7DRV and MSI KT333, will control their own safety measures.

Honestly you put too much "quality" stuff on Intel when one point is ironic, the other is appliable to AMD and Intel.

But back to the price topic, I agree now the price/performance ratio, with 40-50$ more for a better P4 than AMD, is indeed shifting to Intel. But don't expect in a lifetime, that a P4 1.6A would ever touch the XP1600+' current price, a measily but sure as hell nice P/P 55$.

--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!<P ID="edit"><FONT SIZE=-1><EM>Edited by Eden on 09/05/02 04:32 PM.</EM></FONT></P>
September 5, 2002 10:18:28 PM

*sigh*

This is a continuation of our previous argument isn't it (the one that's getting old)? Oh, and if you want links to recent posts, here you go. These are various links for problems that relate to AMD CPU's.

<A HREF="http://support.microsoft.com/default.aspx?scid=kb;en-us;Q261641" target="_new">http://support.microsoft.com/default.aspx?scid=kb;en-us;Q261641</A>

<A HREF="http://support.microsoft.com/default.aspx?scid=kb;en-us;Q256850" target="_new">http://support.microsoft.com/default.aspx?scid=kb;en-us;Q256850</A>

<A HREF="http://www.theinquirer.net/?article=4475" target="_new">http://www.theinquirer.net/?article=4475&lt;/A>

<A HREF="http://support.microsoft.com/search/preview.aspx?scid=k...;en-us;Q321178" target="_new">http://support.microsoft.com/search/preview.aspx?scid=k...;en-us;Q321178</A>

Although Microsoft doesn't say specifically that this involves AMD CPU's (the above 2 links), it's been documented that this kind of problem does not occur on Intel CPU's.

Here is the general search done for AMD CPU's:

<A HREF="http://search.support.microsoft.com/search/default.aspx...
false&numDays=&maxResults=25&withinResults=&Queryl=
AMD+CPU's&Query=AMD+CPU's&QuerySource=gsfxSearch_Query" target="_new">http://search.support.microsoft.com/search/default.aspx...
KeywordType=ALL&T
itles=false&numDays=&maxResults=25&withinRes
ults=&Queryl=AMD+CPU's&Query=AMD+CPU's&QuerySource
=gsfxSearch_Query</A>

Also, just to be fair, here is the search done for Intel CPU's:

<A HREF="http://search.support.microsoft.com/search/default.aspx...
ALL&Titles=false&numDays=&maxResults=25
&withinResults=&Queryl=Intel+CPU's&Query
=Intel+CPU's&QuerySource=gsfxSearch_Query" target="_new">http://search.support.microsoft.com/search/default.aspx...
ALL&Titles=false&numDays=&maxResults=25&
withinResults=&Queryl=Intel+CPU's&Query=
Intel+CPU's&QuerySource=gsfxSearch_Query</A>

Notice how these affect AMD CPU's, but not Intel CPU's. I'll admit, if you search the microsoft knowledge base, then you'll find a few problems for Intel CPU's. But most of the problems with Intel CPU's involve windows 3.1, which had problems with ALL kinds of hardware and software, and also, the other problems involving Intel CPU's are when Windows incorrectly displays a certain stepping. These "stepping" problems have all been fixed, and they do not change the system's stability in any way.

How many times do I have to say this: for the AMD heat protection system to work, you <b>must</b> have a mobo that supports the heat protection, because it is </b>not on-die</b>.

Of course, if AMD made the video, they would try and get the mobo which has the best support for the Athlon heat protection. And the problem is not running the CPU without a fan, but it's when the HSF fails, and also when the CPU gets to a certain temperature, the P4 starts auto-throttling, while the Athlon shuts down at that certain temperature <b>if</b> the mobo supports the protection. With AMD, there are alot of "if's", but with these days, with Intel, you can be sure that it's been tested for stability as well as reliability, and the incorporation of heat protection. Just so you know, AMD's solution is <b>inferior</b> to <b>even the P3</b>. The P3 does the exact same thing as the AthlonXP in terms of shutting down at a certain temp, but the P3's protection is <b>on-die</b>. And anyways, it was very rare to see a P3 running at 60 celsius, so the protection was rarely used.

Also, Eden, if you recall in my post, I <b>did say</b> that the price/performance for the Athlon XP exists *only* below the XP2200+. It's ironic, actually, that AMD's price/performance ends at it's high end CPU, and in terms of price/performance, that's the same place that the P4 starts at (2.2B P4).

Quote:
<i>orgininally written by Eden</i>
Stop spouting FUD, Dark.
Most stable my ass, even if Intel does rigorous testing, while AMD does not as much, please link me to a recent post or world known problem where an AMD proc. started crashing, doing bad mathematical problems. Oh wait, Intel did that! Really, that is just fanboy-expected comments, which you shouldn't have added, because it's FUD and in fact is ironic. In fact, a company called Samson or something like that, started using 100 AthlonMPs in clusters, another company also did similar things for scientific or space stuff. I wonder why Intel wasn't used? Not because Intel has previous problems but because they don't HAVE to choose Intel each time when looking for that so-called ironic stability.

Umm, so what? A company started using the AthlonMP is clusters of about 100, big deal. The reason they (and other compnaies) use the AthlonMP is they need HUGE amounts of processing power, which requires hundreds, if not thousands, of CPU's. If their budget is limited, then they may be buying AthlonMP's because it costs less than a comparable Xeon. Also, that's really the only sector where the AthlonMP is used, in general: the scientific/supercomputing sector of the market. For them, especially in the supercomputing sector, it seems that staying on a small budget is more important than having great stability, reliability, or simply saving space. Also, it wouldn't be a good idea to put those Athlons in racks, since you'd need some powerful, expensive cooling to keep the Athlons running stable, and reliable.

Here's a question for you (and Matisaro, since he's never answered this question ever since I asked it a few months ago):

<b>Why</b> are Intel CPU's <b>favoured and preferred</b> in the server and rack markets as well as the mobile markets (over AMD CPU's)? I'm just very interested in seeing your answer (as well as Mat's).



- - -
<font color=green>All good things must come to an end … so they can be replaced by better things! :wink: </font color=green>
<P ID="edit"><FONT SIZE=-1><EM>Edited by Dark_Archonis on 09/05/02 06:20 PM.</EM></FONT></P>
September 5, 2002 10:19:42 PM

why the hell do you have persian numbering on your sitr? can you even read it?

<b><font color=orange>sing to prolong HDD life; spin right round like a record baby Right round round round
September 5, 2002 10:24:14 PM

lol... did you really have 8733 hits in one day?

<b><font color=orange>sing to prolong HDD life; spin right round like a record baby Right round round round
September 6, 2002 2:07:19 AM

All of those, are SOFTWARE based errors. It is the software itself which is guilty. Just like making Win98 crash, it is not Win98's fault but the program used, which may be badly coded.

Intel had a CALCULATION, Hardware based CPU error in 1994, not sure but almost all those who bought the Pentium at that time had that error. It cannot be fixed because it's an FP problem.
Intel also had released an early P3 1.13GHZ for "comPete's sake"! AMD has yet to release a processor, since the K6 which would by hardware based, with no possible fix, crash or cause calculation errors, which can be very bad if you do Excel or spreadsheet work.

Of course some software may have probs with some CPU, just like some may crash with C3s, others may recognize Intel more. It's all about how the programmers keep in mind which CPUs they are making it for.
That DX8.1 thing, may be apparent, but there is a fix, so again, even here, the point is not about software based errors but HARDWARE.

To me it still seems that this so-called stability thing is moot, because AMD CPUs are as reliable in functioning under a proper software that recognizes all x86 CPUs properly, which is almost 99.8% of programs used today.
Quote:
How many times do I have to say this: for the AMD heat protection system to work, you must have a mobo that supports the heat protection, because it is not on-die.

Exactly my point, NOW AMD is in the game no matter what. If we had this conversation in the early P3 days I could as well have said, OMG Intel has no protection unless you buy a P3 board that complies.
It is obvious you need a board that knows about it.
The diode is on-die btw, and if AMD had done that since day one of K7, it'd have been supported and no fry stories would have happened. Yes AMD was slow, but hell, it's not like they've been producing CPUs as long as Intel, so I'd say they're only 2 years late at most.
My point still stands, AMD now HAS thermal protection if you bought a mobo that supports it. In fact some slip safely with the old thermistor mode.

Quote:
Of course, if AMD made the video, they would try and get the mobo which has the best support for the Athlon heat protection. And the problem is not running the CPU without a fan, but it's when the HSF fails, and also when the CPU gets to a certain temperature, the P4 starts auto-throttling, while the Athlon shuts down at that certain temperature if the mobo supports the protection. With AMD, there are alot of "if's", but with these days, with Intel, you can be sure that it's been tested for stability as well as reliability, and the incorporation of heat protection. Just so you know, AMD's solution is inferior to even the P3. The P3 does the exact same thing as the AthlonXP in terms of shutting down at a certain temp, but the P3's protection is on-die. And anyways, it was very rare to see a P3 running at 60 celsius, so the protection was rarely used.


AMD has a lot of IFs because they just entered that segment. In fact it seems pretty fast how they reacted, since THG. Buy a mobo that supports it, that's all really, especially since you need a Palomino CPU for that anyway. It's just like buying a new CPU, are you gonna go buy a Tbird now, when Palominos and all new AMD CPUs have the protection? Are you to go buy a cheap mobo before June 10th as well if you wanted protection? Point made, since June 10th and the new mobos out, anyone who would buy an AMD would get that protection because he'd buy new hardware, just like someone who would buy a new P4, he'll get that special new technique because the P4 incorporates it.

Whether on or off-die, it has worked, people have given their testimonials, yes it works, even if it's slower off-die. That's all that matters, it's like the clock speed of P4s vs IPC of Athlon, DOES NOT MATTER as long as both output similar performances or balance out.

Quote:
Umm, so what? A company started using the AthlonMP is clusters of about 100, big deal. The reason they (and other compnaies) use the AthlonMP is they need HUGE amounts of processing power, which requires hundreds, if not thousands, of CPU's. If their budget is limited, then they may be buying AthlonMP's because it costs less than a comparable Xeon. Also, that's really the only sector where the AthlonMP is used, in general: the scientific/supercomputing sector of the market. For them, especially in the supercomputing sector, it seems that staying on a small budget is more important than having great stability, reliability, or simply saving space. Also, it wouldn't be a good idea to put those Athlons in racks, since you'd need some powerful, expensive cooling to keep the Athlons running stable, and reliable.


Hmm I thought companies needed absolutly Intel Xeons because they are reliable even if they cost more. Seems not, if people can buy AthlonMPs for big uses even if they are less "reliable" in racks, then obviously it is because they WORK WELL. All my point here was that Intel CPUs are not obliged to be the ones that supply everything corporate below high-end CPUs like ALPHAs, and that AMD is as reliable as Intel, because it has been proven by the buyers.

Quote:
Why are Intel CPU's favoured and preferred in the server and rack markets as well as the mobile markets (over AMD CPU's)? I'm just very interested in seeing your answer (as well as Mat's).

It's a hard question as the one asking why wasn't AMD chosen for the Xbox CPU. Many reasons can come up, some by nature, such as lack of enough supply.
Mobile market, because nobody could make a decent chipset. Not because of heat, otherwise P4Ms would not exist, these things are NOT optimum for lengthy laptop users.
Of course I'd be a hypocrite to say AMD is as known. Intel is a big-time known company, people trust their name because they had experience with them. They could have the same with AMD, but AMD has supplying limits, and very limited worldwide settlements.
In fact only recently did they budge towards HP and Compaq, a small beginning which could propel them to more worldwide knowledge, which can be what AMD needs to finally convince the stubborn companies that Intel is not the only one to make their dreams work. In fact there were news about Hammer being considered for Dells, not to mention AMD is planning a worldwide ADD CAMPAIGN (wow, the day finally arrives?) for Hammer.

That summed up is a fairly good answer to your question which stems a lot of discussion and answers that are possible, but none can be alone to justify.

--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!
September 6, 2002 2:57:11 AM

Quote:
scientific/supercomputing sector of the market. For them, especially in the supercomputing sector

Ya know what floating point math is? AthlonMP's are pretty good at it.

This sig runs too hot.
September 6, 2002 3:03:51 AM

Indeed, I dunno though if Scientific Supercomputing can go far with SSE 2 optimizing, I haven't heard of results with it, but I beleive they are raw FPU hungry, so yes indeed, AMPs for their price, make it a huge saving with lots of positive results.

--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!
September 6, 2002 4:05:18 AM

[avoiding the war]

If you're just gaming, then you don't need the fastest PC out there. Go ahead and upgrade now and buy yourself the best videocard you can afford. Most games are for 450Mhz-500Mhz processors with some levels of some first person shooters requiring 1 Ghz (operation flashpoint for example). Games will not require a 2-3 Ghz computer for quite some time as far as I've seen.

Buy a good videocard, that's what's running the show.

If you are going to be using 3DMax and other CPU intensive programs then I might recommend a different course, but for games there's no reason to wait and wait.

I'm running a 1 Ghz T-Bird. It handles everything. How old is the CPU? I've had it for a year and it's at least 6 months older than that. Anything above a Ghz is gravy as far as I'm concerned. I don't need over 100 FPS and trust me, neither do you.

Enjoy the upgrade :smile:

<font color=red>I'd like to dedicate this post to all my friends, family, and fans. Without them this post would never have been possible. Thank you!</font color=red>
September 6, 2002 4:25:26 AM

I suggest buying the best price/performance ratio if your on a budget right now. Computers usually at max last three years in the gaming world before they will need to be replace because of the newer req, etc. IMO you should NEVER buy whatever is top of the line unless your budget is really big(then get top of the line). I don't give a sh!t how much better the Radeon 9700 performs than my Radeon 8500 because most people(including me) probably won't notice the difference anyways. You should buy something that will do better than average right now, not top of the line. I'd go with an AMD 1900+ right now, because that would offer the best performance/price ratio. Build a system for around $500, and that should last you for at least a year. By that time you will get all the knowledge about the 64bit processors and their non-premium price. Then you could sell your system off on ebay or something and get a brand new system.

P.S. If you are going to get anything top of the line, it should be the monitor, keyboard and mouse. Those you will be using for a very long time.
September 6, 2002 9:16:23 AM

(diving into a trench)

True most games run reasonably at 1Ghz (my PC has a 1.2 TB), but trust me to make the most of the next generation of games (with high detailed physics models et al) such as Doom3, Deus Ex 2, and UT2003 good all round performance is a must - i.e. high memory bandwidth, processor speed etc.

I think the Barton core athlon at either 2800+ or 3000+ is my best option, probably around the time the CH is released, along with a shuttle spacewalker with the AGP slot if/when it is released - BTW can anyone relate any experiences with the spacewalker? How much of a problem is cooling?

Thanks,

Jim
September 6, 2002 9:24:13 AM

Quote:
Why are Intel CPU's favoured and preferred in the server and rack markets as well as the mobile markets (over AMD CPU's)?

Intel cpus are favoured and preffered in the server and rack markets beacuse people who buy and use them are either moron like you or afraid to try new things.

Intel has good mobile chipsets that can drain out last bit of performance from cpus. AMD mobile cpus are paired with very poor performing mobile chipsets. It's the only reason for their bad mobile market share. I prefer AMD over Intel, but if I buy a laptop now I will buy a mobile P-III based laptop
September 6, 2002 11:06:26 AM

What videocard do you have right now?

<font color=red>I'd like to dedicate this post to all my friends, family, and fans. Without them this post would never have been possible. Thank you!</font color=red>
September 6, 2002 1:11:52 PM

"Intel cpus are favoured and preffered in the server and rack markets beacuse people who buy and use them are either moron like you or afraid to try new things."

That's not true. It's because Intel based systems currently offer highest performance. They don't base their purchases on price, they base it on performance/stability, where Intel is currently king.

This sig runs too hot.
September 6, 2002 1:28:34 PM

Single or Dual Athlon MP 2200+ or 2100+ outperforms Single or Dual Xeon 2.2 GHz. AMD 760 MPX is equally stable compared to Intel's MP plarform.
September 6, 2002 1:43:48 PM

In the high end market, brand name is everything. Companies don't have time to build their own systems nor do they have time to troubleshoot every little problem or bother buying from some no-name brand computer manufacturer. Unless AMD can score some major OEM wins for performance computers, they're not gonna get anything done in the high end workstation/server market.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
September 6, 2002 2:44:54 PM

(lol) the number font is random, and the hits are since the beginning of time. That web site only gets attention when I want to learn some new web design techniques. It's been a few years since I wanted to learn anything, web related or otherwise ;) .

_______________________________________________________________________
TeeCee
<A HREF="http://www.digitalfreaks.net/" target="_new">Digital Freaks </A>
September 6, 2002 3:37:54 PM

I've heard predictions that the performance of Barton, with it's 512k cache, may be very close to that of the initial Hammers. If I were you I'd stick with the Athlon, buying a KT400 board, some PC3200 and an inexpensive Athlon, maybe the 2200+. Next year, when the Bartons are cheap, just plop one in and you'll get a nice boost. A system like that should also last you at least a couple years.
September 6, 2002 7:47:43 PM

Hee Hee

You squabbling children are most amusing methinks

Hardware quality & reliability differences between Intel and AMD are negligible, if not nonexistant.

CPU hardware problems have arisen in less than 0.000001% of x86 systems. Don't sweat it.

Never underestimate the bandwidth of a station wagon full of tapes hurtling down the highway
September 6, 2002 7:53:40 PM

Once again LED,
Quote:
stability, where Intel is currently king.

that last bit is FUD. I have provided enough arguments to prove why, and unless Dark_Archonis will reply with valid arguments, I stand uncorrected and undisputed.

Performance-wise, it depends again, if you paid for lower end Xeons, forget it. AMPs at 2200 can be a power house in racks compared to 2.2 Xeons. Although Hyper Threading DOES make a difference there so it shifts a lot. Scientific-wise, not even HT can save the P4s from damnation, Athlons are just aces there.
Some companies do need to save money btw, not SUN-like, but lower end where AMD is targetting. Low end corporate companies need to balance their spending in order to gain revenues, and if you can save a couple of 1000s by paying for AMDs, you WILL get benefit because again, the stability thing is FUD now.

--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!
September 6, 2002 11:44:33 PM

heh i was wondering about that

<b><font color=orange>sing to prolong HDD life; spin right round like a record baby Right round round round
September 7, 2002 1:37:06 AM

Not me really, I am saying what I should agree with you: Quality and reliability is available on Intel as well as AMD. Both are tested, both have not crashed since 2 years (none for AMD though) or had any preemptive releases which might have buggy CPUs, and yes, it is negligible to a point of trying to spot a specific star by pointing your finger towards the sky and expect someone to find the right one when it is behind millions of others, in broad daylight.

BTW why is everyone thinking this is a flame war?
For crying out loud, to me this is a debate and I take it seriously and partly in an aggressive way, but nothing evilish. I like debating, and to me I take this debate with honor, so I hope nobody is thinking this is a flame war, because no, it is a debate to prove who is right or whose arguments stand out.
If someone wants to debate AMD's less-rigorous testing has effects on the core's logic calculation, then by all means go ahead, but you'll be searching far and wide the net to get one result hit, if not being an Intel troll's story to bash AMD.


--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!<P ID="edit"><FONT SIZE=-1><EM>Edited by Eden on 09/06/02 09:38 PM.</EM></FONT></P>
September 7, 2002 2:48:41 AM

Well heres how I stand on the AMD vs Intel debates. About the stability, they are about the same. If you have alot of moneyu (unlike me) then by all means spend your money on top of the line, which would be intel right now. But for me, who needs something like an 1800+ (let me remind that an 1800+ is much much more speed than the average computer user has, probably in the top 90%) Intel wont be able to touch 1800+s performance for the same price. The top of the line proc may change soon when hammer is out, we will just have to see...Intel wins for mobile.

Thread starter, If you dont have enough money for a top of the line system now, I think you should wait a few months for Barton to become widely available, and cheap. Or maybe even wait for hammer.

<P ID="edit"><FONT SIZE=-1><EM>Edited by STEALTHBIG on 09/06/02 10:56 PM.</EM></FONT></P>
September 7, 2002 5:05:54 AM

am bowing ,,,,Eden you da man.

Before i put my ATHLON XP on prozac, it thought it was a PENTIUM 4
September 7, 2002 6:03:02 PM

Ohh, sure blame the software. well, how come Intel CPU's don't have any kind of problems like that? I admit, that Intel has really only had one major problem, but that was for the Pentium a long time ago. And, BTW, In windows 98, there's a way to fix the the FP problem. If you run "msconfig", you can actually enable a workaround of the FP problem. Also, I found out there's an AGP prblem with AMD CPU's and Win2k. <A HREF="http://www.amd.com/us-en/Processors/TechnicalResources/..." target="_new">Here</A> is the link. There are lots of software problems with AMD CPU's, and do you seriously believe that it's Microsoft's fault? Do you seriously think that Microsoft's doing this on purpose? I'm skeptical on that idea. I will agree, for the average user (most people), there is adequate enough testing for AMD CPU's. But in the server/corporate market, these software problems can't exist. Specifically, the DirectX and win2k problems are caused by AMD CPU's themselves, not Win2k or DirectX. Also, I though that AMD had learned from thier past and fixed all of these problems, but the DirectX problem is a very recent one, which tells me AMD still has some room for improvement when it comes to testing. Because of this, that is why I still stand by my argument that AMD's testing is not as good as Intel's. It's close, but not equal, like Mat always says it it. I have searched the net far and wide to find specific info on AMD's testing, and I have never found any information at all. The only thing, really, that I do know is that AMD's validation process for it's MP CPU's is called the "2P validation process". Also, I do know that the MP's are tested slightly longer than the desktop Athlons. Other than that, I have no idea as to what kind of testing AMD does or how rigorous it is. Also, I still have yet to find a reason as to why there are no articles on AMD testing. My theories are that AMD does not want anyone to know that information, or that the testing is pretty dull, and there's nothing extraordinary or unique about it, so no one has bothered to make an article on it. Currently, I believe the latter theory, because it makes no sense whatsoever for AMD to hide their testing. Also, there is a very rare prefetch bug which affects the Athlon XP, but AMD hasn't even addressed it. In fact, I heard the bug is so insignificant that AMD won't bother to fix it. When it <b>does</b> occur, it could freeze your computer, but that's rare. Still, the prefetch bug on the P4 is also rare, yet Intel fixed it, and voila, we see that magical 2% extra performance increase which everyone though was the TLB. In fact, the TLB was always 128 according to Intel, and it seems they were right. There was a bug in CPUID whcih made the TLB display as 64. I believe that has been fixed.

About the P3's, back in those days, did you ever actually hear about heat problems with P3's. I remember that I never heard a thing when it came to heat problems with the early P3's. I'm not arguing if AMD's protection works, I'm arguing that there is no point for it to be off-die, I'm saying AMD should put it on-die. In other words, AMD's CPU's should have FULL, on-die heat protection, and should not need any help whatsoever from motherboards. In fact, many people have pressured AMD for about 2 years now to put heat protection on their CPU's. They finally did, but IMO, it's a lame and dull attempt at heat protection. I mean, you can make whatever method you want for heat protection, but for Pete's sake, but everything on-die! It'll cost very little, and takes up very little die space. Because of this lame protection they have right now, I am turned off by it and it pushes me away form getting an Athlon. If there were enough reasons to get an AMD CPU, I would get one, no doubt. But the fact is, there isn't.

Quote:
<i>Originally written by Eden </i>
Hmm I thought companies needed absolutly Intel Xeons because they are reliable even if they cost more. Seems not, if people can buy AthlonMPs for big uses even if they are less "reliable" in racks, then obviously it is because they WORK WELL. All my point here was that Intel CPUs are not obliged to be the ones that supply everything corporate below high-end CPUs like ALPHAs, and that AMD is as reliable as Intel, because it has been proven by the buyers.

That argument is flawed. If people in the scientific community right now are buying AthlonMP's, it's nesessarily because they "work well", it's because they go for a low price, and they have great FP power, which the scientific & supercomputing community values.

Quote:
<i>Originally written by Eden </i>
It's a hard question as the one asking why wasn't AMD chosen for the Xbox CPU. Many reasons can come up, some by nature, such as lack of enough supply.
Mobile market, because nobody could make a decent chipset. Not because of heat, otherwise P4Ms would not exist, these things are NOT optimum for lengthy laptop users.
Of course I'd be a hypocrite to say AMD is as known. Intel is a big-time known company, people trust their name because they had experience with them. They could have the same with AMD, but AMD has supplying limits, and very limited worldwide settlements.
In fact only recently did they budge towards HP and Compaq, a small beginning which could propel them to more worldwide knowledge, which can be what AMD needs to finally convince the stubborn companies that Intel is not the only one to make their dreams work. In fact there were news about Hammer being considered for Dells, not to mention AMD is planning a worldwide ADD CAMPAIGN (wow, the day finally arrives?) for Hammer.

That summed up is a fairly good answer to your question which stems a lot of discussion and answers that are possible, but none can be alone to justify.

I want to thank you for giving me an honest answer. this is my point: Intel is a known and trusted brand in that market. Intel CPU's are known for great quality, reliability, and ever since the P3 coppermine came out, they are known for low power, low heat solutions. AMD does not have that reputation: they still have to earn it. Hopefully, AMD has learned something since the release of the Athlon, and we will see if that's true with the Hammer. I agree that P4M's still have a way to go, but it's obvious that P3's currently are king in the rack and mobile markets. Also, here's a very interesting link form overclockers about a heat and power comparison of AMD & Intel CPU's. I highly suggest that you read it. Mat, if you're reading this post, I highly suggest you also read this article:

<A HREF="http://www.overclockers.com/" target="_new">http://www.overclockers.com/&lt;/A>

From there, go to "CPU reviews and tips", and once you're in that menu, go to the ""CPU Die Size - The Cooling Challenge Ahead" article, and read it.

BTW, one of the main reasons that an AMD CPU was not used in the Xbox was because the power drain would be too high, and a good HSF would be needed to cool the Athlon. Also, Intel has some sort of deal going with Microsoft.

Finally, let me explain to you where my "stability" argument gets it's roots. In the server/corporate/mobile market, Athlons tend to run quite hot, since the are turned on alot, and alot of strain is put on them. Generally, they run hotter than Northwood's and definitely hotter than P3's. As you know, the hotter your PC (and CPU) runs, the more stability is compromised. That's why in the server/mobile market, it's very important to have a cool-running CPU, that doesn't drain too much power.

LED, I know what that is, and I stated that point in this post. Also, do you know what an Itanium 2 is? Do you know that it can cream an AthlonMP in Integer, as well as FP operations (running in 64-bit). And, of course, you probably know that an Itanium 2 puts IBM's Power 4 to shame in the SPEC CPU2000 benchmark.

Spitfire_x86, you're a total moron. People buy Intel CPU's over AMD in the server and rack markets because of all the above reasons that I listed. And AMD's chipsets are not as stable as Intel's. Intel tests it's server boards for <b>10 000</b> hours. That's pure, raw testing time. I <b>highly</b> doubt that AMD tests it's mobo's for that long. Imgod2u also added some valid points.

It seems that I figured out that most techies and enthusiasts base their purchases on price/performace. Most ignore everything else. These same techies probably all buy Samsung instead of Sony, or Hyundai instead of Toyota/Honda. Probably, these techies would even rather buy a Ford, rather than a Honda/Toyota. At least there are a few techies that base their purchase on overall QUALITY.


- - -
<font color=green>All good things must come to an end … so they can be replaced by better things! :wink: </font color=green>
<P ID="edit"><FONT SIZE=-1><EM>Edited by Dark_Archonis on 09/07/02 02:22 PM.</EM></FONT></P>
September 7, 2002 10:01:12 PM

Quote:

Ohh, sure blame the software.

Well let's take it from the programmer's side. The guy is new, he makes these sloppy codes. BAM, Win98 crashes.
So now we blame MS?
Nope, it's the guy who did the prog. Same thing for CPUs, lack of AMD recognition puts programmers on Intel's side.

I and almost 90% here have had or currently own AMD CPUs, and I seriously doubt any of them had SOFTWARE based problems such as the DX 8.1 problem.

Again we were debating HARDWARE based CPU problems and this is where I consider your stability argument especially since you added it to someone who is a home user, FUD.
Quote:
Finally, let me explain to you where my "stability" argument gets it's roots. In the server/corporate/mobile market, Athlons tend to run quite hot, since the are turned on alot, and alot of strain is put on them. Generally, they run hotter than Northwood's and definitely hotter than P3's. As you know, the hotter your PC (and CPU) runs, the more stability is compromised. That's why in the server/mobile market, it's very important to have a cool-running CPU, that doesn't drain too much power.

Again under many circumstances, the difference in temps is not too big. Xeons, which are more powerful than NWs, output more heat, in the end either Xeon or MP heats up. It's up to the IT guy of the company to know what to set up, and it's up to AMD to set good Retail coolers. So whatever comes Retail with AMPs MUST comply to MP server systems.

I am not in the mood to reply to other things, I am over my head with so many things to do now, but I will try to read the article and re-read this post to reply to other arguments...



--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!
September 8, 2002 5:30:27 AM

im going to inject something here, dont flame me but just a thought, one of the reasons INTEL is so popular, and not AMD, is all the advertising INTEL does. everytime you turn to a chanel, you here that intel jingle. LOL AMD needs to use that mc hammer song (hammer time) for their hammer processor.

Before i put my ATHLON XP on prozac, it thought it was a PENTIUM 4
September 8, 2002 6:26:39 AM

Perhaps commercials would help. But Intel has over 30 yrs in the business. They established themselves w/ their products, not commercials. Most of those years no one could do what they do. Intel has had that large share of the market for YEARS, and I doubt a MC Hammer gimmick would take much away. All eyes are on Hammer right now. It could either be worth its weight in gold, or flop. AMD has the best marketing for the tech community. It's all hyped up, everyones watching, now let's see it.

This sig runs too hot.
September 8, 2002 6:54:25 AM

You should do a little research on the history of both Intel and AMD. If you did you'd know that Intel is only about six months older than AMD. You'd also know that both companies have very intertwined histories and that it's mainly by luck and litigation that Intel emerged as the larger, more successful company. That AMD has survived and developed into the company they're today in spite of all of Intel's shenanigans is amazing. So I say F' Intel!

I like the Pentium IV, I really do! And it's so versatile. You simply won't find a more stylish or decorative key chain ornament or paperweight.
September 8, 2002 7:54:11 AM

Actually, if you followed your own advice, you'll know that Intel was a semi-conductor to manufacturer microprocessors first long before AMD, which started out as merely a semiconductor manufacturer, not designer, begin to create MPU's. AMD then began to design flash memory and other parts along with being another semiconductor manufacturing plant like TSMC is now. They would sell of their fab space to other companies (one of which was Intel) to make chips for them (since Intel couldn't meet all of the demand back then). Only around the time of the P5 core did AMD start designing their own MPU and by that time, Intel had gained a great deal of the marketplace. So yes, AMD is almost as old as Intel, but they started in a different market and entered the microprocessor design game a lot later.

"We are Microsoft, resistance is futile." - Bill Gates, 2015.
September 8, 2002 8:02:34 AM

So you're telling me something I don't already know? Thanks anyway!

I like the Pentium IV, I really do! And it's so versatile. You simply won't find a more stylish or decorative key chain ornament or paperweight.
September 8, 2002 1:48:39 PM

When AMD did the train add back in 1999, supposedly it created a huge demand they could not supply, since then it was because of that, as one of the reasons, that lots of OEMs don't choose them, because they could not honor the demand.

So yes AMD CAN make adds that change people's minds, you know commericals are almost subconscious drivel!

--
When buying an AthlonXP, please make sure the bus is at 133MHZ, or you will get a lower speed!
September 8, 2002 2:12:21 PM

AMD got creative after the purchase of Nexgen. AMD was just an Intel clone during the 80's, early to mid 90's. Intel was by far the innovator in processor technology, w/ AMD benefiting from cross patent agreements. They were around during the same time, but Intel had the superior product. And it wasn't till 1995 when AMD released the AM5x86 processor that they were even considered competition, which was mainly used to upgrade old pentium75 systems. What you're spewin is AMD fanboy propaganda......I read your post in another thread, you got it out for Intel. Stay awhile and learn somethin.

This sig runs too hot.
September 8, 2002 4:23:48 PM

Hehehehe...ummm, some people here might find your history lesson useful, I suppose. I've supported myself from working in this industry since the late 80's and I'm more than just a fanboy. I buy and use nothing but AMD machines in the businesses where I work, from laptops, to desktops and even servers, I avoid Intel like the plague. So I'm not just an enthusiast.

Ok, I'm ready to learn something..... ;-)

I like the Pentium IV, I really do! And it's so versatile. You simply won't find a more stylish or decorative key chain ornament or paperweight.
!