• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Phenom/Barcelona Reviews/Previews and Pre-release Discussion

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

]-[itman

Joined
Sep 24, 2001
Location
Arizona
You'll understand when you read it

http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=3092&p=4

Keep in mind they do say with faster memory, barcelona should extend it's lead a bit further. So, it looks like intel still has the lead for the next product cycle, atleast on the desktop, although now AMD is atleast pretty competitive. Let's cross our fingers that they can get the speeds up by Q1 to match intel and be able to raise their asp to start making a profit again. Enjoy.

*edit*
Two things to point out:
1) The platform they are using still seems to be pretty buggy (see page 3), this may also degrade performance a bit.
2) Atleast on the server/workstation end, AMD seems pretty confident as they sent anandtech a Xeon server at 2.33ghz to compare to the 2ghz Opteron server.
 
Last edited:
Damn, their site must be getting hammered. Never seen Anandtech load so slow. :santa:
 
Ooo phenom... rendering is not your strong point.

Zvisuel.

Notice that the Intel CPU has the advantage when it comes to raw processing power: it is about 19% faster in a single CPU configuration.

3DMax 9

In other words, the new quad-core is about 20% more efficient than the previous dual-core generation, core per core, clock per clock. This is also shown by the score of the quad Opteron 880 (2.4GHz). But it is not enough to beat the Intel armada: even a 2GHz Xeon will probably slightly beat the current 2GHz Opteron 2350.

Fritz

Chess programs are branch intensive integer applications, and although the Opteron 2350 is a bit better (clock for clock) than the previous generation of Opterons, it is the Xeon "Clovertown" that comes out on top. The quad-core Xeon is 30% faster.

Mind you actually the xeon runs at 15% higher clockspeed.

Come on more benches there must be more in it.

At least finally next year I am going to have the cooler and more silent proc.
 
At first I could read it normally but by now pageloads take a minute.

16-21% improvement over K8 in games is good, and it's gonna be better with future games, 25% clockspeed 16% in games.

Does anybody know how C2D scales ?

That's a server hence no fans over the heatsinks, the processors are in the path of the airflow ( much higher pressure than desktops of course usually much louder ) therefore it needs no fans. It depends on the design some cases have huge 12cm+ fans behind the drives to suck air between the hotswap scsi drives push out at the back of the case, the noise as like running an aircon.
 
These are server, of course they are running with fanless heatsink.

I wonder what type of heat these Barcelona's give off fanless. Can Intel's Quad Xeon's run fanless? I know they would run dang hot if they can!
 
I wonder what type of heat these Barcelona's give off fanless. Can Intel's Quad Xeon's run fanless? I know they would run dang hot if they can!

for the new Xeon G0s they can run fanless on stock speed and voltages, because they can handle up to 85C
 
Assuming the preview scaling is correct (I believe the jump from K8->K10 will be more around 20% on average on desktop once a proper platform is setup), then looking here and calculating for a Phenom, it would actually put a K10 cpu at a virtual tie with a Core2 cpu in Oblivion and about a 4% advantage in HL2 episode 1.

Then, if you look at K10 mhz scaling versus Core2 scaling on the same page, you will see that a Core2 (3ghz versus 2.33ghz) shows a 13-14% performance increase with a 28% clock increase, whereas a Phenom would give a 16-17% performance increase with only a 25% clock increase. Now, if you take those results and let's say I'm right and K10 will scale even better on a proper platform, then with good yields it will probably become the gamer's choice cpu, though it will probably be more about money at that point. Of course this is also only considering 2 games, more benches would be needed to draw a firm conclusion, but these 2 games have been known to be pretty responsive to cpu's.

*note: I hope this is clear, I'm running on less than optimal amounts of sleep right now :)
 
Last edited:
That is pretty good. Although performance only matches up to 2.33ghz, they kill intel on the efficiency scale. Fanless, no FBDIMM, split power plains, less power consumption when idle. That is something intel will not be able to fix anytime soon although they offer better performance above 2.33ghz which AMD cannot compete with.This also makes the AMD at 2ghz a better buy than intel because FBDIMM and fans etc. are more costly. So far for the server market it looks very promising.
 
I think going from 2.33 to 3G the vga is the bottleneck so it's doesn't worth to compare how K10 scales from 2G to 2.5G.

I didn't think of that, so I tested a bit more, found another example,
http://anandtech.com/cpuchipsets/intel/showdoc.aspx?i=2963&p=8

You can see using 2.93ghz and 2.4ghz Core2 cpu's, you see about the same scaling as with the other cpu's. You could argue that you hit a graphics bottleneck in HL2 here as well, but it doesn't fall very far outside of the other examples. You could also make a pretty good case that given the server platform, the graphics performance might not scale as well for the barcelona system either.
 
Fanless, no FBDIMM, split power plains, less power consumption when idle. That is something intel will not be able to fix anytime soon although they offer better performance above 2.33ghz which AMD cannot compete with.This also makes the AMD at 2ghz a better buy than intel because FBDIMM and fans etc. are more costly.

Fan fan fan...
As I said before it's normal for servers not to have fan Intel's 4x4 (16 core) 2U server doesn't need cpu fans either even at 2.9G. ;)
 
I didn't think of that, so I tested a bit more, found another example,
http://anandtech.com/cpuchipsets/intel/showdoc.aspx?i=2963&p=8

You can see using 2.93ghz and 2.4ghz Core2 cpu's, you see about the same scaling as with the other cpu's. You could argue that you hit a graphics bottleneck in HL2 here as well, but it doesn't fall very far outside of the other examples. You could also make a pretty good case that given the server platform, the graphics performance might not scale as well for the barcelona system either.

Okay we will see when they finally release something.

Until then.

If we are to believe that 15% is the best we'll get on average, taking into account that Penryn is around 5% faster than Conroe, the updated architecture from AMD alone isn't enough to really compete with Intel. In other words, price matters.
 
Wow I didnt know that..I wonder what heat those 4x4 servers produce at 2.93ghz. LOL I actually really dont care. Anyway I want to get me a hydrojet when it comes out . http://www.hardwaresecrets.com/article/437 Im just happy that they are doing better in the FP section once again. That is a nice niche they can keep all to themselves.

I probably wont upgrade till next year so hopefully they will have better chipset drivers and etc. to help increase the performance in the desktop market for everyday applications. If not Ill just wait for penryn or get a q6600 overclock the snot out of it. Im not really a die hard fan of AMD or intel.
 
Last edited:
Same here, definitely no upgrade on the AMD rig till this goes to 45nm, and then I decide who has better price performance.
When Neha comes out penryn might go cheap and upgrade the intel first.
 
Last edited:
Back