AMD 965 - where have i heard that number before?

Status
Not open for further replies.

dragonsprayer

Splendid
Jan 3, 2007
3,809
0
22,780
AMD in another genious move names their new cpu the 965!

Let me think. that sound familar? oh ya, the 965 p4! wait the 965 chipset! wait the new i7 965!

Its bad enough when your cpu's fail to deliever as the phenonm did, but the new phenom II are a bargin - note to amd!
note: do not copy intel names to promote a really good cpu, come up with your own name!


you guys miss me?

intel fan boy #1: IFB#1 you guys miss me?
 
Oh, you mean because Intel would never copy AMD by something like using x86 64-bit extensions (especially when Itanium was going so well), use an integrated memory controller, or try and make a dedicated graphics card. Oh, wait a minute............
 
Oh, this IS nothing new, that 3700 AMD? Yep , it was named after the P4 equivilent of its needed clockspeed per performance.
So, when they were ahead, they did it then as well, so, your point?
 


Actually AMD is letting Intel use x86-64 and thus making us stay with x86 which is old and needs to be thrown out but change is near impossible.

And no Intel is designing a new GPU based with their own ideas. AMD just bought ATI and is taking the credit for plans that ATI had before AMD purchased them (ATI had up to R800 aka the HD5K series out before AMD bought them.), big difference.

Now if AMD actually went out of their way to create a new GPU of their own design along with chipset then maybe Intel would have copied them. But since they took the easy road and bought ATI its not intel copying its Intel making the playing field even so to speak.

And Dragon, seriously who gives a crap with the naming? If it performs for those who buy it it works. All it is is a way to limit the OC potential of it.
 
Hey, DS, love your new models, especially the mouse

cubancomp.jpg
 

xaira

Distinguished
wasnt l3 cache amds idea too, no ones gonna mistake a $250 chip fora $1000 chip, although, i as a fervant amd supporter must hang my head in shame of the 140w tdp.
 

theholylancer

Distinguished
Jun 10, 2005
1,953
0
19,810
if this was named SPARTA and not that shitty sempron.......

imagine this, the sparta proc (this 965) holds its own againest I3 and i5 for a little while, until intel unleashes price cuts to them and the mobo prices goes down and the 300 (lets face it, who pays for a top of the line proc when your OC overhead is small) Spartans are slaughtered and AMD goes back and does R n D.

then AMD releases true K-10 and pwns intel, with a 3 to 1 ratio!!!

then it'd be lawls

damn that could be fun
 


LOL - you're far too cruel :). Besides, with a 140W TDP, that iron could be powered by 965 X4's :D.

BTW, interesting how AMD, despite giving the world X64, fares so poorly on overclocking on 64-bit OSes:

A 625MHz overclock is nothing to sneeze at, but it certainly does not come near those 6GHz~7GHz clock speeds we see on a regular basis with the Phenom II series. Granted, reaching those high clock speeds requires LN2 cooling but there is another important reason. We have discussed it several times and still do not have an acceptable answer from AMD about the inability of the Phenom II to clock much past 4GHz with a 64-bit operating system. Even with LN2 cooling we have not successfully benched past 4.4GHz with a 64-bit OS.

Once again, we tried XP 64-bit, Vista 64-bit, and Windows 7 64-bit and the results are always the same. As we near 4GHz, the voltage requirements increase dramatically and the clocking ability of the processor decreases in much the same manner. This does not occur in a 32-bit operating system, which happens to be the recommendation for any sort of benchmarking activities with the Phenom II.

:bounce:
 


But that doesn't appear reasonable, seeing as how the superleaky P2's can oc to 7GHz on a 32-bit OS. Seems far more likely that AMD didn't do as good a job with critical timing paths in 64-bit mode as they did with 32-bit mode.
 
No. First off, the super leaky parts oc beyond anything Intel currently makes, and those arent for sale, even tho Intel tried to outdo it with their own super leaky 975
Secondly, at normal operating temps, which were talking about here, and not those other chips, temps for stabilization are more important on AMDs current lineup than Intels.
Read the reviews, on every one, the thermals are whacked using a 64bit mode, so its obvious.
Correct that with a new stepping, and the problems solved
Also, in earlier oc reviews, those same chips never reached the AMD presntations, or the TWKR levels, came close but fell behind to around the Intel cherry picked level
 


So why don't the i7's also show a big thermal increase in 64-bit mode?

Sorta seems like AMD did the pooch on their claim-to-fame X64, at least with the P2 :D
 
Here is what I found in a quick search:

Higher Overclocks in Vista 32-bit vs. 64-bit?
In preparation for this article we also wanted to test one more theory: that Phenom overclocks better under 32-bit Vista than 64-bit Vista. The basis for this theory is that the number of active registers increase in 64-bit mode, which creates additional workload routines that leads to increased transistor switching that ends up increasing thermals and loads on the IMC. Well, you get the picture, not exactly a cycle of events that would lead one to believe an already clock starved core design has a chance when moving to a 64-bit environment, well at least Vista 64-bit. In case you're wondering, Intel CPUs seem to overclock the same regardless of OS but just for kicks we tried our 9850 under the two OSes to see how things differed:

Of course this is a year old - no P2, no Nehalem. However since the P2 is basically a P1 on 45nm, I'd guess at least half of the the above remains true :).
 

theholylancer

Distinguished
Jun 10, 2005
1,953
0
19,810
umm, if this can be fixed by a CPU revision, chances are they are not present and I want to see proof of Ph II doing this.

Because a die shrink don't mean they didn't do under the hood work, they do work on their CPU all the time, which is why we have CPU revisions of the exact same chip, and some revs work better (IE the D0 vs C0 deal, intel, but AMD does these revs too)
 


The "proof" was in the original comment I made, some 8 posts above, concerning the Anandtech P2 965 review. AT found the 965 hit a wall around 4HGz on 64-bit OSes - XP-64, Vista-64 and W7-64. Same as the 955 and IIRC the 945.

So it looks like despite a process node shrink & I dunno how many steppings, this problem hasn't been fixed since P1 on 65nm a year ago.
 
This speaks nothing of the entire usage , nor the amperage.
It appears the volts can be higher on the 32bit OS, which could mean that since the entire core osnt being used, a higher vcore is allowed, whereas that isnt so when fully loaded.
Since AT failed to take power usage draw, and only listed voltage in, it means nothing, and proves nothing, other than theres a difference, which is what we knew before.
Think of it this way. If 50% is being used, and you upped the vcore by 50%, it maxxes out.
But, using 100%, you cant up the vcore. So again, unless they showed power draw, they got nowheres in this "investigation" other than to have AMD look bad, and its all on their Phail
 

theholylancer

Distinguished
Jun 10, 2005
1,953
0
19,810
can anand or someone repeat the test but with LN2 and hefty voltages about this ceiling, take a chip that is proven to run at 6 Ghz with LN2, get it run at 6 Ghz with LN2, boot into 64 bit with stable 32 bit settings and see what happens?
 
Look at this table. The top oc. If thermals are controlled , it shouldnt make any difference right?
http://www.xtremesystems.org/forums/showthread.php?t=213248
The link shows the top oc was done on phase, which handles the thermals, same as LN2 would.
I said earlier, theyre tricky with their thermals, and using a 64bit on the same cooling system mind you, itll be too much.
However, take that away as lancer suggests, and we see the outcome from my link, tho only using phase, it appears its enough, and so too would be a stepping change
 


It seems AT already did - see this from the first post of mine above:

Even with LN2 cooling we have not successfully benched past 4.4GHz with a 64-bit OS.
 
Status
Not open for further replies.