TCCD at 2-2-2-5 250+ Mhz

[man_utd]

wow (http://www.xtremesystems.org/forums/showthread.php?t=44527)

[SteveOCZ]

Its nothing more than a die change. Everyones TCCD should do about that soon....depending on PCB and some other things, but you get the idea

[man_utd]

Wow, if that is the case, can you give a time frame on soon>

[SteveOCZ]

Basically anything being made from now on will be.... depending how much stock the company has it could be longer before the new chips filter in.

[man_utd]

Will these chips need as much voltage as the one posted at xs? At that much voltage, it seems bh-5 has been reborn.

[SteveOCZ]

http://www.xtremesystems.org/forums/showthread.php?t=44647

[man_utd]

Thanks for the link, can't believe I skipped over that, ahhh, bh-5 has been reincarnated.

[Reefa_Madness]

The modules pictured in post #133 of the XtremeSystems thread linked below are of the Mushkins reviewed by PCStats, which have a mfg date of 431, so anything after that date (approximately Aug 1, 2004) should be from the new revision. It would be safe to assume that the high volume retailers are going to get rid of their old stock quicker so they would be the safer bet to have new stuff, possibly even the PDP stuff, since it was later to come to the market, would be a good bet.

http://www.xtremesystems.org/forums/showthread.php?t=44527&page=6&pp=25

Doesn't Everest (formerly AIDA32) and some other software report the mfg date of ram? That would be a way of finding out without having to take the spreaders off.

Edit/Correction:
The manufacturing date info is diplayed by CPU-Z (ver 1.24) in the SPD tab. I thought it was available in AIDA32, but now can't locate it so I was probably just confusing the two tools.

[flapperhead]

The modules pictured in post #133 of the XtremeSystems thread linked below are of the Mushkins reviewed by PCStats, which have a mfg date of 431, so anything after that date (approximately Aug 1, 2004) should be from the new revision. It would be safe to assume that the high volume retailers are going to get rid of their old stock quicker so they would be the safer bet to have new stuff, possibly even the PDP stuff, since it was later to come to the market, would be a good bet.

http://www.xtremesystems.org/forums/showthread.php?t=44527&page=6&pp=25

Doesn't Everest (formerly AIDA32) and some other software report the mfg date of ram? That would be a way of finding out without having to take the spreaders off.


good observation.. also its good to remember these modules could be an anomaly. i experienced something similar with some simpletech nitro pc4000. (later revision chips that only responded to 2.9 volts max)a while back.
both sticks maxed out in the low 290's at 2.9 volts. having tested alot of the later hynix chips i knew 2.8/2.9v to be about their max.. however for some reason one of the sticks just picked up and flew (310 mhz)when i increased the voltage to 3.2.

[Reefa_Madness]

good observation.. also its good to remember these modules could be an anomaly. i experienced something similar with some simpletech nitro pc4000. (later revision chips that only responded to 2.9 volts max)a while back.
both sticks maxed out in the low 290's at 2.9 volts. having tested alot of the later hynix chips i knew 2.8/2.9v to be about their max.. however for some reason one of the sticks just picked up and flew (310 mhz)when i increased the voltage to 3.2.

I don't think that they are unique.

One of the OCZ guys even stated that before long everyone's TCCD would be this way because of the die revision. All of the new TCCD will be based on the new revision. They don't do much between 2.8-3.1, but take off after that.

Bet you that sales volume of DDR Boosters will go up once this becomes common knowledge. I recently ordered an A64 3000 (939) and a MSI K8N Neo2 Plat to play with. One of the reasons I chose that board was because that mobo will play nice with a Booster and I wanted to see what bh-5 would do. I may have to break down and buy some Samsung in a few months (next year's budget) and try it at 3.4 (Booster's max on that board).

Edit:
It just occurred to me that I may not have to wait. I've just received my RMA exchange last Thursday of 2 x 512 OCZ PC3200 Rev 2 so it may very well have the new chips, plus I'm waiting on an RMA from PQI. Wouldn't that be sweet if as a result of the originals not working, I ended up with 2 gigs of the new stuff to go with that new A64/MSI mobo!

Not a bad swap for just the cost of shipping. I can only hope.

[situman]

ok took the heatspreaders off my pqi and the sammy chips have the "431" at the end. Seems like the more voltages I give it the more errors when i was still using hte ocz booster. I did have a fan over the modules

[austinbmxnig]

i would love the thought of getting over 250+ with tight timings, definately checking back in this thread as soon as more info is found

[man_utd]

Situman, how did you get the tape to come off?

[Styyn]

ok took the heatspreaders off my pqi and the sammy chips have the "431" at the end. Seems like the more voltages I give it the more errors when i was still using hte ocz booster. I did have a fan over the modules

Situman, do you have a dmm to check your vtt with? Lately I've become rather leery of the OCZ Booster due to the way some mobo's track vtt. If your board happens to be like mine it will set the vtt via your bios vdimm setting and not by what your ram is actually getting. I.E. you set 2.8 vdimm in bios so your vtt gets set to 1.4. Now you crank the Booster up to 3.3v, but your vtt is still down at 1.4 and suddenly you're unstable.

Just a thought.

Maybe an OCZ guy can speak to the general vtt tracking specs across the huge market of mobos? Is this a widespread problem? Do most mobos need to be vtt modded for the Booster to work properly or do most track vtt dynamically or "on the fly" by checking the real volts the dimms are getting?

[Reefa_Madness]

ok took the heatspreaders off my pqi and the sammy chips have the "431" at the end. Seems like the more voltages I give it the more errors when i was still using hte ocz booster. I did have a fan over the modules

Were you going over 3.1 or did you stop at 3.0 or 3.1 because you were getting no results. If I remember correctly, in one of the threads over at Xtreme, it was stated that between 2.8 and 3.1 the additional voltage did not help and in fact, hurt performance. It was only after you exceeded 3.1 that the additional voltage resulted in increased performance.

You might consider going back and re-testing with the Booster at levels in excess of 3.1v to see if in fact it does allow you to run 2-2-2 at faster fsb.

[situman]

i went as high as 3.2v and thats when i started getting the errors at 260fsb 2.5-3-3-8

[Reefa_Madness]

Thank for the reply.

That is interesting because the chips used in the ram reviewed by PCStats were dated 431, like yours. It is possible that there were some of both revisions manufactured during that week, I guess, or that they were producing both, but that is less likely.

We will have to keep an eye out for mfg dates to try to get a better idea of when the change took place.

[man_utd]

Reefa, where in everest do you see the chip information?

[SteveOCZ]

Maybe an OCZ guy can speak to the general vtt tracking specs across the huge market of mobos? Is this a widespread problem? Do most mobos need to be vtt modded for the Booster to work properly or do most track vtt dynamically or "on the fly" by checking the real volts the dimms are getting?

From what I have seen it's only Abit boards that do not track VTT properly. Asus and the rest seem to have used a different method :clap:

[Banyan]

this is silly to me... i mean the only way you are going to see a difference as opposed to running these at low volts and a tad looser timings is if you are a benchmark junkie correct? lolz...

[situman]

True I cant really tell the difference anymore at such a high speed. Nothing really stresses the system except for games, but thats really cause the 9800pro is lacking (imagine anyone saying that about this vidcard 6 months ago?).

[Reefa_Madness]

Reefa, where in everest do you see the chip information?

Well, good question. I was sure that it was shown there (AIDA32), but when I went to find it in order to post an answer for you I had no luck. I don't know if I am overlooking it or more likely, I confused my facts.

I did go to CPU-Z (ver. 1.24) and found the information in the SPD tab so I'll go back and edit my prior post to reflect this.

Thanks for asking the question and making me look it up.

[austinbmxnig]

that mushkin pc3200 lv2 version2 512x2 looks perfect for me + ddr booster

[situman]

Well, good question. I was sure that it was shown there (AIDA32), but when I went to find it in order to post an answer for you I had no luck. I don't know if I am overlooking it or more likely, I confused my facts.

I did go to CPU-Z (ver. 1.24) and found the information in the SPD tab so I'll go back and edit my prior post to reflect this.

Thanks for asking the question and making me look it up.

u mean i didnt need to open my heatspreaders?

[Reefa_Madness]

u mean i didnt need to open my heatspreaders?

I checked out the date and such on a rig with some Samsung PC2700 ram that I have in an XP rig. It is the only rig I currently have running that has a current version of CPU-Z. I have not checked to see just what all you get from a TCCD chip using ver 1.24, but you will, for sure, get the mfg date. Based on what it told me about the PC2700, I would suspect that you will get enough info to verify that it is TCCD without having to take the spreaders off.

When you get a chance, download ver 1.24 and check out the ram and post what you got. I am in the process of having wood floors installed in my home and the room with the PCs is next so I will not be posting for a few days unless it is from work. If I get a chance, I'll try to swap out ram and see what some of my TCCD displays in CPU-Z.

Actually, I'll go do that now and post results in a little while.

[Reefa_Madness]

Well, how is this for crap...with the the PC2700 it gave me the ram's serial number and the mfg date. With the OCZ PC3200 Plat Rev 2 (TCCD) is would not display this info, the captions and boxes for the info were just grey'ed out. That sucks. I checked it on an WIN 2K system, somebody with XP needs to try it out and see what results you get. Is it possible that because the ram is new it is not identified by CPU-Z?

So much for my easy way to tell.

[man_utd]

The information should be read from the spd, so it should not have anything to do with CPU-z being out of date. It might be, but I doubt it.

What brand was the pc2700 you had? Perhaps some brands grey out the info, while others do not.

[Reefa_Madness]

The information should be read from the spd, so it should not have anything to do with CPU-z being out of date. It might be, but I doubt it.

What brand was the pc2700 you had? Perhaps some brands grey out the info, while others do not.

It was Original Samsung PC2700 2 x 256 sticks that it read correctly. I then swapped out for the OCZ with the TCCD chips and it didn't show anything (gray'ed out box). Not sure what to make of it. It did identify the ram as being OCZ and with the correct part # so you're right, it is reading that from the info on the ram.

Can some of you guys try out different ram and see what kind of results you get? I don't know which versions show it and which don't, but I was using ver 1.24 with the PC2700.

[mattspalace]

My Corsair 3200XL also has greyed out boxes. I have a set of Mushkin V2 that should be here in a couple of days, so I'll see what those say..

[mxa]

Can some of you guys try out different ram and see what kind of results you get? I don't know which versions show it and which don't, but I was using ver 1.24 with the PC2700.

I have the PQI Turbo 3200 (2,2,2,5) with windows XP Pro and the manufacture box shows all '0's and the Part Number, Serial Number and Manufacturing Date are all blank.

[Reefa_Madness]

I wonder if it only shows it for "original" branded stuff like my Samsung PC2700. I have some "original" Hynix PC2700 that I will have to try. Does anyone have some original branded PC3200 that they can test so that we have something other than 2700. If only the "originals" show the date, then that feature is worthless for what we need.

I am so sure that I've seen this feature in another program but right now I'm drawing a blank as to which one. I'll just let it search the Memory Banks and hopefully it will be located.

[man_utd]

On my laptop (in sig) I can see all of the model information, it's samsung pc2700, and all the information is viewable. I will go and check my mushkin level 1 and pqi in a sec.

[man_utd]

Nope, neither say.