• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

amd venice 3500 and 3800 preview!

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Originally posted by: Acanthus
After "giving them some time" it looks like the 3500+ and 3800+ overclock very similarly to the FX-55. They will however be cheaper, so while there is a marked improvement, i dont see anything really exciting about it. The chip reacts poorly to voltage increases once it hits the clock wall as well.

I don't think you've given them enough time yet, the CPUs haven't hit retail, we still don't know what the 3000/3200+ venice will do. AND, we still don't know about if the chips can do 4x1T with doublesided ram. Even if they're no better overclockers than the best Winchesters, if they have such an improved memory controller the upgrade to venice would be worth it alone for many. Being limmited to using 2 sticks of ram is not fun.

On the flip side if the memory controller still can't handle 4 sticks, but the cheapies can get to 2.8GHz on air and possibly more, the upgrade should be worth then as well - as it shouldn't be hard to sell off a tried and true Winchester.
 
Originally posted by: Duvie
Originally posted by: Acanthus
Originally posted by: Duvie
I think you are judging too early and haven't really looked at the numbers in how they relate to the FX55...We are talking about improvement with the venice cores, not mind blowing improvement...If it allows us 200-300more mhz at stock that is something.


There are examples already of veice cores on stock cooling and phase cooling, stock vcore and overvolting....just look at the numbers now and look behind final stability at the moment. None of these uys are tweaking yet. they are all holding high ram dividers and aggressive timings...no one is even trying right now...they are seeing how it boots. I think 2.95ghz so far from a 2.2ghz still booting into windows and running superpi 1mb (far better then runnig snadra scores) at stock vcore is impressive and not many of the FX55's did this....

But this is with $1000 cooling...



I know!!!

Howvere in one of the instances it was done with stock cooling....

IN another instance we can look at it comparatively to an FX55 and phase cooling to get an understanding how much better this is versus the FX55...in terms it is relative how much better an venice core air cooling and stock vcore will do versus the FX55....

YOu can look beyond the cooling at the moment....

A K8 beats Iintel at 2.4 Ghz ( well most of the time), you can do all you want with the put downs the cores not hitting 3 Ghz on air, but who said they would ? and do AMD really need to hurt intel in the performance world ? Think not.
 
If we'r goona go off the first run of silicon then your missing the point.

yeilds in the early day of Prescott were very low (3.4,3.6Ghz) were like gold dust, now a year on and there cranking 3.8's out with no problems.

Give it time, tweaks and then maybe, just maybe you'll get 3 Ghz/3.2Ghz on air.
 
Actually there are not many FX55's that did 3.0ghz at default vcore as well..This is 2 examples and they both did it......Most of us are happy hitting a 2.5-2.6ghz average now with our 3000+ to 3500+ winchesters....Our expectations were that some may hit 2.9-3.0ghz...I think that looks like that may happen still....Until the 3500+ is tested on air we dont know if it would have made it itself...
 
pc ice hurt his back pretty bad and wont be able to post the water and air results right now, but he will post them later, hope he gets well soon...

btw, it seems there are already venice cpus in stock in 2 shops in europe! 😀
 
Originally posted by: saaya

sorry, i should have mentioned it, yeah it was just a tease, its just the first results we started with 🙂

3ghz was the max, all benches could pass at that speed, we didnt have time to test prime though since we're in a hurry to test other stuff first 😀

it went fine with 3.1ghz even, but no benches at that speed were possible...
pc ice is doing a great job so far, tomorrow oppainter himself will borrow the chips and put it under his cascade 😀
from -40 to -100 😀 hehehe

and no, the cooling pc ice used doesnt cost 1000$, its a single stage and he also builds and seels a few of those, i think they cost around 400$, i dont know for sure though 🙂

i have no idear why dave didnt up the vcore some more... well we still have a few days to play with this cpu, so im sure we will find its sweet spot 😀
and btw, the 3800+ did better than this one, 3.45ghz with 1.68v under a mach2gt wich is worse cooling than the single stage pc ice used for this 3500+!

now we're getting somewhere....
 
update of water results, the chip does 2.8ghz with 1.4v vcore (stock) on water 😀
not too god, but not bad either...
cooling was asetek waterchill

higher vcore barely seems to change anything though... lets hope that not all chips dont like higher vcore settings and that its an oddity of this chip...
 
So are we saying the venice core is clock for clock faster then the winchester core as well???

Do you have any proof that the venice's pipeline was lengthend?? We have a shite talker here who thinks it has and is the reason the are starting to scale towards 3ghz....
 
Back
Top