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

Sabertooth 990fx GEN3/R2.0

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.
It's funny, I was surfing the egg today and noticed there isn't one Gen3 available at their site ATM. This means either they sold them all( no message saying out of stock) or Asus pulled them back for some tweaking. Maybe to address some issues or put a sticker on the box say ing 5.0 FX ready !
 
It's funny, I was surfing the egg today and noticed there isn't one Gen3 available at their site ATM. This means either they sold them all( no message saying out of stock) or Asus pulled them back for some tweaking. Maybe to address some issues or put a sticker on the box say ing 5.0 FX ready !

The Canadian egg doesn't even have the Gen3 listed and given that AMD doesn't appear to be selling the newer FX processors to Joe Enduser, I wonder if they've been yanked to fix some underlying issues.

Cruising the forums at a couple of other sites seem to indicate a problem with the onboard sound as well although I never experienced that.
 
Agreed, I blew a 970 board myself last week just trying to do a small voltage OC to my FX-6300, can't imagine how it is for an FX-8350 :sly:

Actually an 8320 will work the mobo harder at any given clockspeed than an 8350 as the lower binning requires that you use even more voltage / current to keep it happy.
 
Assuming that it even is working harder? I have seen plenty of (and this isnt AMD, I know) 2700K's have higher voltage than 2600K's and I would assume AMD works that way as well. There is PLENTY of crossover between bins. Is there more to it than that on the AMD side? It really all depends between those two CPUs (and even Intel's CPUs).
 
An average sample 8350 can uaually hit 5Ghz on 1.5-1.52V, of the 4 8320s I have run through my system all of them required 1.6v+ to be stable above 4.6Ghz. Thats a big voltage jump and raises the current draw quite a bit. Of course you could argue that its easier to provide 200W at 1.6V(125A) than at 1.5V(133A), but that does not account for the increased current draw from the increased voltage and higher heat output. Either way the 8320s suck power if you want to shoot for the stars.
 
you guys all seem to be right.
I assume that all vishra processors start life as 8350's and then are tested out to 83xx-63xx and 43xx and for the 6 and 4 core processors the failed cores are just told to go to sleep.
am i correct in this?
 
Ya basically thats it. The cores are not necessarily bad but they are flawed enough that they require far to much voltage when compared to the other cores on the same die. Disabling 2 cores for a .2v drop baseline voltage reduction makes good sense. Sometimes inevitably the cores are probably legitimately dead though.
 
so Im on the right track with that.
the other interesting thing here in amdland is that the processors that need the higher vcore and run hotter seem to overclock higher, they get much less work done at the same time.
take my 4170, it'll clock to the moon, it'll take 1.8 vcore and is harder to cool than my 8350 at clocks above 5.0. I have no idea on earth why this is.
 
well poke me with a knife and I leak a little too...
but whats "leakage" as it pertains to my processor?
does it just define a **** poor performing processor that clocks real easy?if you bought a desktop with this 4170 in it you would take it right back because it is just a dog.
 
Last edited:
Electron leakage is similar to electron migration. Its basically electrons escaping the circuit. Leaky transistors tend to be more tolerant of abuse, but also dont always provide consistent results. Those lost results are were the performance hit comes in as it causes parts of the instruction to be re run. Theres a boatload of more info on that if you google for it.

Start here http://www.anandtech.com/print/1611 search for the term "leak" it will take you about half way down. Theres another really good article on this but I cant seem to find it right now.

Heres a non overclocking article on leakage, <Warning excessive scientific information inside>
 
Last edited:
That's why I cal mine a little piggy C_D. The more you give it the more it wants, cause half of it's running down it's face.
 
This I can confirm first hand as I was there yesterday when I picked up a R2 Sabertooth ..... Canada computers has in stock and on the shelf the new Gen 3 Sabertooth boards. Pretty much all stores have them showing in stock. At the same time there are very few of the R2 Sabertooths left .... Not sure if this is Asus drying up old stock or people grabing them while they can. Just 2 left in Ottawa now out os 5 stores.
 
I have found some reading on "leakage", I have a hard time wrapping my head around it, but it seems we use high wattage at such low voltage.
 
This I can confirm first hand as I was there yesterday when I picked up a R2 Sabertooth ..... Canada computers has in stock and on the shelf the new Gen 3 Sabertooth boards. Pretty much all stores have them showing in stock. At the same time there are very few of the R2 Sabertooths left .... Not sure if this is Asus drying up old stock or people grabing them while they can. Just 2 left in Ottawa now out os 5 stores.

When I returned the second Gen3, my local Canada Computers showing 2 in stock and indeed they did have 2; both open box and I'm pretty sure one of them was the first one I returned the prior week!

Over on the Non-ROG Asus AMD forums there are a bunch a people having problems with on board sound as well.

I still maintain that the Gen3 has serious issues and that ASUS has pulled it until they can work them out.
 
Back