Sign in with
Sign up | Sign in
Your question

AMD Phenom X3 CPU's are out at newegg

Last response: in CPUs
Share
April 19, 2008 10:29:07 PM

are there any reviews of them?
April 19, 2008 10:29:41 PM

Sweet but how did they end up with 3.5 L2 instead of 4
Related resources
April 19, 2008 11:01:20 PM

reconviperone1 said:
Sweet but how did they end up with 3.5 L2 instead of 4


yeah, i guess the wallet got a bit tight, although that would mean that it is not just failed yield,
April 20, 2008 12:27:48 AM

reconviperone1 said:
Sweet but how did they end up with 3.5 L2 instead of 4


I'm guessing that is total cache 512KB x 3 and 2MB L3 wichh is 1.5MB L@ total and 2MB L3.
a b à CPUs
April 20, 2008 1:16:05 AM

yep.

The price needs to go down that is for sure.

April 20, 2008 1:46:31 AM

Headline: AMD Phenom X3 CPU's are out at newegg!

So is the Q6600!
April 20, 2008 2:19:14 AM

It's really living up to it's name...Tripple Cripple.
a c 100 à CPUs
April 20, 2008 2:30:46 AM

Quote:
^ good to see you back Skittle

Its expensive cause it just released, just wait 2 weeks and it'll be cheap.


My guess is that it's expensive because AMD has good yields on the quads. The triple-cores are just quads with one core fused off, so if they have to blow fuses on a perfectly good quad to meet the demand for X3s, then X3 prices will be nearer to those of the quads. AMD doesn't want to intentionally have to take a hit on their revenues just because the demand on certain CPUs isn't optimal.
April 20, 2008 2:32:09 AM

i thinh i might jes buy one :o  amazing price/performance!!!

::sarcasm::
a b à CPUs
April 20, 2008 2:38:38 AM

If they drop the price I can see they would be a good rig for encoding ... a cheap workhorse machine ... so you can spend more time on the main rig gaming ... LOL.

I will wait till the price drops a bit.
April 20, 2008 5:15:58 AM

blackpanther26 said:
I'm guessing that is total cache 512KB x 3 and 2MB L3 wichh is 1.5MB L@ total and 2MB L3.

LOL, duh forgot about that my bad
April 20, 2008 5:18:54 AM

Reynod said:
If they drop the price I can see they would be a good rig for encoding ... a cheap workhorse machine ... so you can spend more time on the main rig gaming ... LOL.

I will wait till the price drops a bit.

Me three, get it me three(i know its lame but i couldnt resist
a b à CPUs
April 20, 2008 6:03:00 AM

lame .... lol .... clever pun there.
April 20, 2008 6:59:03 AM

I just think it's the retailers that is why they seem so expensive but they'll go down with in a day or to.
April 20, 2008 10:12:39 AM

Maybe if we all bought AMD they might have enough moolah to actually create a decent chip...





On second thoughts, Intel is by far the better manufacturer atm. EVERYONE BUY INTEL
April 20, 2008 1:50:27 PM

Those numbers don't look like an epic fail. They just need to lower the price. And the review site needs to upgrade their AMD testbed to something that is AM2+. Every single one of those processors was clocked higher than 2.3ghz except for the e4500.
April 20, 2008 2:42:35 PM

Quote:
Newegg prices are always outrageous when the product first arrives....
and those reviews were horse crap.........something was wrong there......we all saw the review months ago that showed the tri's were nearly as fast as the quads in multi-threaded apps......so something was definately wrong with those recent tests.



How can tris be nearly as fast as quads in multithreaded tests? It's a 3:4 ratio. And they charge high prices for new items everywhere, not just Newegg. Nothing's wrong here.
April 20, 2008 2:49:08 PM

I find it interesting that the tri-cores are still considered to be a part of Phenom 9000 series, which so far has been reserved for AMD's quad cores.
a c 100 à CPUs
April 20, 2008 2:52:03 PM

dagger said:
How can tris be nearly as fast as quads in multithreaded tests? It's a 3:4 ratio


It depends on how many threads a "multithreaded" application spawns. If it spawns two compute-intensive threads, any chip with 3 or more cores will have similar performance. The two program threads will run at 100% on two cores and then the OS's background tasks will run on the other core or cores. Most of the programs tested in consumer-type sites are games, which tend to have one or two heavy threads and then little else to run on the CPU.

Even programs that have many compute-intensive threads may not run much faster on a quad than they do on a triple-core chip. It all depends on the memory bandwidth available and needed as well as cache usage and other system resource contention issues. A good case to demonstrate this is running SPECfp_rate on a dual Clovertown Xeon system. Performance scales well up to four threads if the OS schedules the four threads to run as two on one CPU and two on the other CPU, so the FSB bandwidth is utilized optimally. But once you cross the 4-thread barrier, the FSB starts to get hammered something fierce and you get much smaller increases in performance when scheduling more threads, even when there are idle cores available.
April 20, 2008 3:11:54 PM

MU_Engineer said:
It depends on how many threads a "multithreaded" application spawns. If it spawns two compute-intensive threads, any chip with 3 or more cores will have similar performance. The two program threads will run at 100% on two cores and then the OS's background tasks will run on the other core or cores. Most of the programs tested in consumer-type sites are games, which tend to have one or two heavy threads and then little else to run on the CPU.

Even programs that have many compute-intensive threads may not run much faster on a quad than they do on a triple-core chip. It all depends on the memory bandwidth available and needed as well as cache usage and other system resource contention issues. A good case to demonstrate this is running SPECfp_rate on a dual Clovertown Xeon system. Performance scales well up to four threads if the OS schedules the four threads to run as two on one CPU and two on the other CPU, so the FSB bandwidth is utilized optimally. But once you cross the 4-thread barrier, the FSB starts to get hammered something fierce and you get much smaller increases in performance when scheduling more threads, even when there are idle cores available.


x264 scales near linearly with # of cores available. A mild 3.0ghz Q6600 will blow the pants off one of those crippled phenom's.
April 20, 2008 3:50:50 PM

Uhm Okay, those reviews are CRAP. They are using a regular AM2 board, not AM2+. I dont trust those reviews, wait for more credible ones to come. A few weeks ago there were ones showing it being a great buy. They definately need a AM2+ hell I think the best thing about the Phenoms is the HT 3.0
April 20, 2008 5:09:45 PM

^ so what about all of AMDs promises of "compatibility" and "drop in replacement" ... I thought that (and the whole 4 core thing) was the best thing about phenom/barcelona.

I for one applaud anyone who shows how the phenom performs on the AM2 platform... Its user base is far greater than AM2+
April 20, 2008 5:55:01 PM

But, the performance differs, yes the chip will work flawlessly in a AM2, and thats what AMD was saying, and thats true. But the true performance of a Phenom comes out in a AM2+ board.
April 20, 2008 6:15:35 PM

Quote:
Newegg prices are always outrageous when the product first arrives....
and those reviews were horse crap.........something was wrong there......we all saw the review months ago that showed the tri's were nearly as fast as the quads in multi-threaded apps......so something was definately wrong with those recent tests.

Newegg reviews are horsecrap because idiots who dont even own the product post their opinions instead of fact they have from owning the the product. Their were 4 reviesws for dual core sempron, and every one was to dissuade someone from using it, but none of the revieweres owned it.
April 20, 2008 7:31:18 PM

Xpyrofuryx said:
But, the performance differs, yes the chip will work flawlessly in a AM2, and thats what AMD was saying, and thats true. But the true performance of a Phenom comes out in a AM2+ board.


HT 3.0 does nothing for performance, since Phenom is hardly bandwith limited even on HT 1.0. HT 3.0 might make a difference for multi socket CPU platforms, but desktop CPUs aren't even saturating HT 1.0. Its like throwing an extra 8 lanes to a freeway during off peak tmes, if you'd like.

April 21, 2008 2:09:24 AM

epsilon84 said:
HT 3.0 does nothing for performance, since Phenom is hardly bandwith limited even on HT 1.0. HT 3.0 might make a difference for multi socket CPU platforms, but desktop CPUs aren't even saturating HT 1.0. Its like throwing an extra 8 lanes to a freeway during off peak tmes, if you'd like.


Fairly true, you'll only really see HT3.0 show an improvement with Heavily threaded apps, or those that have heavy memory usage.
April 21, 2008 2:48:46 AM

I haven't found any Phenom HT speed comparisons yet, but heres one for an Opty 146, and it shows how little effect HT speed has on performance:

http://eclipseoc.com/index.php?id=6,47,0,0,1,0

Even assuming Phenom doubles the bandwith requirements of an Opty, based on these results HT 1.0 should still have bandwith to spare.
April 21, 2008 3:25:43 AM

well apperantly AMD did not get HT3.0 right with the Barcelona chips so hope they can fix it with their 45nm chips.
April 21, 2008 1:28:51 PM

Mathos said:
Fairly true, you'll only really see HT3.0 show an improvement with Heavily threaded apps, or those that have heavy memory usage.


You mean, in other words, the very things they're supposed to be testing, yes...?
April 21, 2008 10:04:24 PM

Quote:
Thanks for explaining that. From the earlier benchies we saw, the tri's ran nearly as fast as the quads. These benchies showed completely different results. Which one do you believe?


What earlier benches are you referring to? Do you have a link?

From all the (p)reviews that I've seen, such as these two: http://www.erenumerique.fr/test_processeur_amd_phenom_x... / http://www.pcgameshardware.de/?article_id=622354 the only time a tri keeps up with a quad is in non multithreaded apps (as you would expect). In multithreaded apps it performs about as well as can be expected, roughly 3/4 the performance, somtimes slightly better than that, as even multithreaded apps don't always show perfect scaling with core count.
April 22, 2008 6:34:37 PM

Why are they deactivated now? WTF?
April 22, 2008 6:43:40 PM

Quote:
Honestly the Q6600 at 2.4 is only very slightly faster than a Phenom at 2.4. Its the pure clock speed from the overclockability of the Q6600 that makes them in 2 different leagues.



Or, you could run it at STOCK SPEED and handily beat the Phenom.

I don't understand why every person who loves AMD wants to underclock the Q6600 when comparing it to AMD products.
April 22, 2008 6:46:11 PM

Another thought: The box says "Triple Core Design"

I think that is kind of misleading. It's actually a quad-core design, but it has a defect. It wasn't designed to be three cores, it was comprimised.

So the box could more accurately read: "Triple Core Compromise"
April 22, 2008 6:54:17 PM

Huh, they are deactivated. Epic fail to deliver?
a c 100 à CPUs
April 22, 2008 7:57:20 PM

TechnologyCoordinator said:
Another thought: The box says "Triple Core Design"

I think that is kind of misleading. It's actually a quad-core design, but it has a defect. It wasn't designed to be three cores, it was comprimised. [sic]

So the box could more accurately read: "Triple Core Compromise"


Both Intel and AMD have sold many CPUs with parts disabled- cores, cache, FPUs, other features. The GPU makers do that also- the 7900GS, 8800GS, x1900GT, x1950GT, etc. That issue has been rehashed so many times on this forum that it's not even beating a dead horse any more- it's beating the grease spot on the ground where the dead horse used to be.

The "triple-core design" bit could very well refer to the fact that AMD designed the quad in such a way that one of the cores could be easily disabled by fusing off if faulty, leaving a fully functional 3-core chip with all of the memory interfaces, crossbar, and L3 cache of the initial quad-core die still intact.
April 22, 2008 10:08:27 PM

Don't [sic] me! You pretentious little....

JK, jk, you're a wealth of knowledge. I still don't think it qualified as a "Triple Core Design", but I see your point.
a c 100 à CPUs
April 23, 2008 2:54:09 AM

TechnologyCoordinator said:
Don't [sic] me! You pretentious little....


Just seeing if you were paying attention :D 

We all make typos and miss keys when typing on keyboards, especially crappy laptop keyboards- I was just giving you a hard time.
April 23, 2008 4:54:15 AM

Back to the original topic, I just noticed that both of the Phenom X3 processors are now listed as "deactivated item" at Newegg. I wonder if this might have been an attempt on the part of Newegg to determine the proper pricing for these chips.
a b à CPUs
April 23, 2008 10:34:14 AM

Like the Celerons and all of the other Intel lines that run slower eh??

Seriously you don't believe the slowest Intel CPU and the fastest are in some way different (for either AMD or Intel)... other than defects or fuzed off cache that prevent it from running flat out by way of adjusting the multiplier or FSB rated speed?

Surely we have had this hammered home so even the most stupid user knows the basics ... or do we simply write you off as a n00b??

How then is the Triple core any different from an E6600 - both are downgraded in speed, the E6600 has the full cache, and the triple has a core fuzed off.

Pick any CPU other than the fastest full cache unit in a given line ... are they all failures??


Lets see ... using your "crippled logic" ...

Desktop processor specifications
Processor NumberΔ Cache Clock Speed Front Side Bus

45 nm
Q9550 12MB L2 2.83 GHz 1333 MHz Good CPU
Q9450 12MB L2 2.66 GHz 1333 MHz (speed crippled)
Q9300 6MB L2 2.50 GHz 1333 MHz (speed crippled, cache crippled)_

65 nm
Q6700 8MB L2 2.66 GHz 1066 MHz (speed crippled)
Q6600 8MB L2 2.40 GHz 1066 MHz (speed crippled)


If you can't grasp these essentials you shouldn't be posting in the CPU section ... you should be simply READING.
April 23, 2008 10:39:32 AM

MU_Engineer said:
Just seeing if you were paying attention :D 

We all make typos and miss keys when typing on keyboards, especially crappy laptop keyboards- I was just giving you a hard time.



LOL. If anyone deserves a hard time, it is me! I have got to get the spell check fixed at work!
!