OCZ 3200 Plat Rev 2 sucks?

joker_927

n00b
Joined
May 30, 2005
Messages
40
I doubt anyone of you guys remember me from the DFI venice oc thread over in the OC section, but it took me about a week and I have finally found the sweet spot of my CPU: 290x9 1.5V

Since I have a 3000+ I dont have a large multiplier, so that meant I needed to raise the FSB a lot. Because I knew all this before I bought my memory I spent a good $65 MORE on the OCZ platinum revision 2 twin pack. It looks awesome, but that's about where my happiness with them end.

@ stock they are supposed to run at 200mhz 2-2-2-5 2.7V

I cannot get them to run at stock speeds because any multiplier below 3/4 and windows wont even boot. This seems to be a common problem with dividers that allow a huge difference between fsb and mem speeds. But! I can still use a 3/4 divider, which at my FSB yeirlds 217mhz. I thought OCZ PLATINUM would be able to do 217mzh at same timings, but nope. I cant even get into windows. just reboots. So I raised the voltage, 2.8, 2.9. 3.0 3.1, even 3.2 freakin volts wont even boot into windows.

ok, maybe my memory is just really picky and won't do 2-2-2-5 at anything over stock. So I set it to 217mhz still and played around with timings. I havent tried every variation but the best timings I got were 2.5.-4-3-7. THAT IS HORRIBLE for only 217mhz. And remember I am trying voltages from the stock 2.7 up to 3.2,

There is the wierd part, I can run the memory all the way up to a 9/10 divider. 1:1 won't work even at 3-4-4-10 and 3.2V, so DDR 580 is out the picture) but at 9/10 I get 261mhz which is a nice jump, but the lowest timings that I can get to pass memtest are 2.5-4-3-8.

with the 5/6 divider I get 241mhz and the max that will pass memtest are 2.5-3-3-7 and I forgot the voltage.I think its 3.0V

Did I just get some horrible sticks?
 
How can you expect it to do anything amazing when they speedbin sooooo many different speed ratings?
 
wow, that's an insane amount of voltage to be giving tccd.. i wouldn't be surprised if you partially killed it :(

and it'll clock best with 2.5-3-3 or 2.5-4-3
 
(cf)Eclipse said:
wow, that's an insane amount of voltage to be giving tccd.. i wouldn't be surprised if you partially killed it :(

and it'll clock best with 2.5-3-3 or 2.5-4-3


Yeah, Samsung and I gurantee you DDR500 with 2.5-3-3 ;)
 
(cf)Eclipse said:
unless the 3.2v killed it :(

People do good with TCCD 440 with 3.1v. I'm sure 3.2v is fine for temporary, one-time usage.

Dont ever go to 3.2v again. Its for Winbond and Micron only.
 
Then new OCZ Rev2 plat (v1.1) is not TCCD, its TCC5. That would explain the shitty OCing.
 
Dillusion, heh! Post at the same time with the same info! LOL!

Good job man. :rolleyes:
 
Dillusion said:
Then new OCZ Rev2 plat (v1.1) is not TCCD, its TCC5. That would explain the shitty OCing.

TCC5 goes in their best memory though. The PC5000 uses TCC5. At least that's my understanding...

edit: Oh, only up to 4800. TCC5 isnt BAD stuff, per se. It's just not great. TCCD is good memory, but its not the best for an A64 anyway so no big loss.

second edit: They discontinued the PC5000. I'm almost certain it had TCC5 though.
 
Ok, i think my memory is bad.

seriously

cause the ONLY way I can get into windows is 3-4-4-10 @ 3.0V

Why 3.0V? Cuase 2.7,8 won't load windows, 2.9 loads windows but reboots as soon as I open a program.

Can anyone say RMA?
 
joker_927 said:
Ok, i think my memory is bad.

seriously

cause the ONLY way I can get into windows is 3-4-4-10 @ 3.0V

Why 3.0V? Cuase 2.7,8 won't load windows, 2.9 loads windows but reboots as soon as I open a program.

Can anyone say RMA?

Something else is wrong. OCZ handtests their memory. They have to with their speed binning process.
 
Could a setting on my board be wrong?

I dont know what "hand testing" or "speed binning" is.
 
basically a speed bin.. test parts for how fast they can go..
there's probably quite a few levels, one that picks up the fastest chips that can do 300mhz, another for 275mhz, 250mhz, 225mhz, 200mhz, ect.. and probably some timing levels within there..

basically, the higher the bin, the "better" your ram is


and if you get a hand picked stick of ram, basically it means that you have the very top stick from the respective bin, which means that it's very nice.
 
Dillusion said:
Then new OCZ Rev2 plat (v1.1) is not TCCD, its TCC5. That would explain the shitty OCing.

Most people are getting really good clocks on it. Alot are able to hit 270-300Mhz+. It clocks pretty much the same as TCCD except Winbond just tests it at 240Mhz instead of 250Mhz but that doesn't mean it wont clock the same or better. It just means your chances aren't always as good.
 
burningrave101 said:
Yea thats what i meant. I been viewing too much Winbond information lately :p.

Winbond is very interesting. I'd like to know more about them, but Crucial memory is in my future anyhow. Cant say I'm disappointed by this Winbond-style Infineon ram, though :D
 
HybridHB said:
My OCZ R2 does 250 mhz at 2.5-3-3 at 2.6v. Maybe its dfi's problem?

Maybe, if he's not using the samsung optimized bios for it. But that speed sounds about right since that's what your memory is pretty much rated at. Well, if it were TCCD. Stupid TCCD/TCC5.
 
I had trouble getting very high clocks out of my OCZ TCCD with my new Venice chip. It seems quite a few people have. I dont know if its just a weak memory controller or what but there does seem to be a problem with reaching high clocks on TCCD memory even after doing alot of tweaking with Rev E chips.
 
I loaded bios defaults and the memory runs fine @ 200mhz 2-2-2-6 @ 2.65V

Darn, this means they work.
 
robberbaron said:
Something else is wrong. OCZ handtests their memory. They have to with their speed binning process.

HAHAHAHA!! Just laughing at OCZ, OK now that is out of the way.

joker_927 said:
Since I have a 3000+ I dont have a large multiplier, so that meant I needed to raise the FSB a lot. Because I knew all this before I bought my memory I spent a good $65 MORE on the OCZ platinum revision 2 twin pack. It looks awesome, but that's about where my happiness with them end.

I would agree!!!

Just out of intrest have you tested each stick separately? I also have problems running 1:1 above stock 2-2-2-5 at 2.85v, an the amout of vDimm makes no diffrence...

joker_927 said:
I loaded bios defaults and the memory runs fine @ 200mhz 2-2-2-6 @ 2.65V

Darn, this means they work.

No RMA, sound like we both are in the same boat.


THIS IS JUST MY $.02
 
Assuming your using a DFI NF4 board....


Make sure your using the new bios (3.10p i think.) This version adds full support for TCCD and TCC5 ram. Previous versions were fubar.

Dont try anything faster then 2.5-3-3, it aint gonna happen.

Check out OCZ's forum. They have a guide on how to set up all the crazy setting in the DRAM options on DFI NF4 boards based on fsb ranges.


I had mine running at 250fsb 2.5-3-3-8 2.8v or 2.9v. They got pretty fucking hot tho under load. If u really want to push em you might want to get a XP90/120 to cool em down.
 
well the guys at OCZ are the ones that recimended that i use 3.10. oh well
 
Back
Top