It beats the P4 3.73 GHZ EE in Doom3 :eek:
Printable View
It beats the P4 3.73 GHZ EE in Doom3 :eek:
yeah but at 2,5mhz a dothan is almost at the max clock at air....
Dothan does. Dothan on the adapter would be lucky to beat a 1.8GHz chip in D3.Quote:
Originally Posted by Sheik
Hehe. FX55 all the way :D
Quote:
With the CT-479 ASUS has effectively demolished all other desktop Pentium M solutions; there's no reason to even consider a 855GME motherboard from AOpen or DFI, the ASUS solution is cheaper, better performing and is even a much more stable overclocker. Kudos to ASUS for a job extremely well done with the CT-479, it's the only option we'd recommend for those interested in a desktop Pentium M system.
That being said, despite being paired up with enough memory bandwidth, the Pentium M continues to fall behind in desktop performance. As a gaming platform and as a general purpose/office machine the Pentium M does fairly well, but it is in content creation, workstation and media encoding applications that the Pentium M continues to fall behind in. Part of the problem is that the Pentium M needs clock speed to compete, which we saw when we overclocked it up to 2.56GHz. But even at 2.56GHz the Pentium M wasn't a competitive CPU when it came to tasks like media encoding, indicating that if the Pentium M is to succeed on the desktop it's going to need some architectural improvements.
Edit: Okay there is no problems with gaming performance with the adapter in the Anandtech review: http://www.anandtech.com/cpuchipsets...spx?i=2382&p=7
Can you pinmod a voltage increase using this thing?
I wouldn't choose PM for low power because of the price. A64 is a better choice (and 67 W for a 3500+ is still just plain BS).Quote:
Originally Posted by WiCKeD
PM 7x5 : 21 W @ 2100 MHz max
PM 7x0 : 27 W @ 2267 MHz max
T64 ML : 35 W @ 2400 MHz max
T64 MT : 25 W @ 2200 MHz max
mobile A64D : 35 W @ 2000 MHz max (1.35 V)
(Ok you never know if those numbers are made for CPU's that never will show up, just like that 67 W number seems to be.)
That A64D number can easy be translated to desktop CPU's (1.4 V):
3500+: 41.4 W
3200+: 37.6 W
3000+: 33.8 W
...and then there's some headroom for undervolting if you feel like doing that. Much cheaper, more availability, more mobos, and A64E seems to be even more undervoltable = THE SFF choice.
I'm surpised that the new AOpen mobo haven't showed up, one store here in Sweden says (yeah..) they will have it on April 11, thought it would show up earlier in the US.
well, someone please slap that adapter on IC7 now, please :)
Anandtech just reviewed this adapter on an Asus board:
At the end of the article he concluded the P-M was good for everyday desktop tasks and gaming, but it lacked in multimedia and encoding performance. He managed to OC the 2.13 dothan to 2.56ghz (multis become locked when Dothan is used on desktop, and vcore as well). Stock multi of 16 and increased the FSB from 133 to 160.
Just to give you guys a hint at performance, it beats a FX-55 in Doom 3 with High Quality settings. WOW
Click here for complete review.
no bios settings, but it can be changed in windows.Quote:
Originally Posted by Tenknics
so no probs as the clockgen is also available.
...only at 1024x768, with 0.3 FPS, and the FX-55 at stock speed... :stick:Quote:
Originally Posted by Tenknics
that is interesting how it beats the FX-55 in some tests but can't even beat regular a64s in others. Intel has some interesting artitecture here, should be tweaked a bit though with some more content creation, workstation and media encoding instructions because that is where it's seriously lacking.
TBH I question the anandtech numbers as they contradict and differ quite alot from previous numbers posted about the dfi board, their 2.56 dothan is paling ni comparison to other dothans at same speed, plus I was under impression that if you set graphcs to high that its the gpu that takes the brunt not the cpu so that skews numbers even more.
Edit: turns out the reason anandtechs numbers were different is 'cause they have a newer bios that now adds vcore and multis :rehab:
Of course there is a new Beta BIOS from Asus USA for the P4P800-SE offering V-Core up to 1,95 Volt an best gaming performance.
Also the magic FSB200 limit has broken:
http://www.legitreviews.com/images/r.../dothanfsb.jpg
yeah babiiiiie !!!!
Hey guys this is Nate over at Legit Reviews. I just wanted to let everyone here at Xtreme Systems that I have updated my article since I first posted it.
I first tested the board with the ASUS 1008 BIOS revision, but late today ASUS sent me Beta 1009 and it fixes the graphics performance issues, adds voltage support, and the available memory dividers now work. (Before the 1:1 divider didn't run 1:1).
I have spent some more time this afternoon re-overclocking the system and the results thus far are great. I'm running 1.6V on the core with no volt mods with the available options in the new BIOS.
Here is a screen shot of some of the recent testing I have been doing...
http://valid.x86-secret.com/show_oc?id=8208
http://www.legitreviews.com/images/r.../dothanfsb.jpg
I have some of the new BIOS images and more details on my overclocking on this thread.
Thanks for the update Nate! I've updated our front page story with your posting :up:
220FSB on a Dothan, can't wait to see what these chips can do on the right platform! :hehe:
You have to realise that Dothan has 2MB of cache. Its cache also has the lowest latency out there (yup, even lower than A64). Thats why its able to perform very well in apps that benefits from large cache (low latency gives it additional boost). In apps that don't benefit from large cache, its low latency also helps in boosting performance.Quote:
Originally Posted by bh2k
However, Dothan does not have SSE2 nor is the architecture optimised for video apps. Dothan's FPU isn't exactly powerful either (P4 has SSE2 and sheer clockspeed while A64's has the most powerful FPU of any x86 CPU). Have to take note that Dothan is afterall a mobile CPU. If Intel is going to cramp everything P4 has into Dothan, its power consumption will not be at where it is now.
Ok it looks like price wise (since were all going to overclock and put it on water or phase anyway), the dothan adaptor kicks some serious ass. They are going to release a bios for the p4c800-e as well which is great news. So a volt modded board with the vdroop and vmods should allow any voltage issues to be handled, put it under some ln2 or cascade and lets see what they can really do! One question i do have though is which chip would be the one to go for. I.e. a 770 533fsb / 765 400fsb and why? Would a lower spec chip give more???
how do you plan on doing that, the adaptor blocks 2 of the 4 mounting holes :(
great job nate =)...
lol :) what forum is this? do you seriously think for minute that blocking two of the holes will stop the ingenuity of the people on this forum? If so try the pc-world forum :stick:Quote:
Originally Posted by XeroHouR
Getting a 400fsb chip may actually be better for overclocking due to the higher multiplier. Damn man do you have 4 VapoLS's?Quote:
Originally Posted by oublie
the problem is that all 400 fsb chips are all of older stepping and not capable as high fsb as 533mhz ones.. :(
guys with adapter what multis are unloked????
All multipliers from the default one downto 6x are usable. :)
Quote:
Originally Posted by aoc007
No 2 ls' , waterchiller and a 3 stage cascade soon, once i sort some things out :rehab: