looks like also BH-5 will have a revival!!!

[Bluebeard]

about Winbond UTT CH-5 we can find thread here and already posted quite alot about...

BUT:

today i have been talking to our technical engineers and also took a close look on the price list and what did i see?

Speed Premium DDR450 with BH-5 in addition to the Speed Premium DDR400 CH-5.

Hope to get some samples asap to tell ya the results ;)

[EnJoY]

Where are they getting it.

[esdee]

Now thats more good news from twinmos!

[tictac]

TwinMOS rocks :)

:toast:

[setyotomo]

agreed :toast:

[EnJoY]

Did they just have this BH5 all along or are they making it again themselves?

[perry_78]

It'd be interesting to know! While TCCD can do high FSB, BH5 can do comparatively lower frequencies with tight timings! :D

Why did Winbond exit the RAM market anyway?

[Ulti]

Damn, I guess I'll have to sell my UTT then :stick:
Bring on the BH5 action :toast:

[celemine1Gig]

...Why did Winbond exit the RAM market anyway?

Perhaps only to reconquer the market with flapping flags after some absence. :D :banana:

[craig588]

Why did Winbond exit the RAM market anyway?

There is not enough money in it for them.

[iddqd]

TwinMOS rocks :)

:toast:
What are you talking about, TwinMOS is foolish.

Not all BH-5 is good BH-5. In fact, there is still a lot of BH-5 chips out there on the market. However, most of the "overclocking" weeks have already been bought up, and the remaining chips are the kind that people get and whine "buhuhuhoo my BH-5 doesn't go over 220MHz halp meh!", and then wonder why. Of course it's a combination of things - sh1tty PCB, bad week, lack of any binning out of bad chips... cheap materials. And TwinMOS fits into all of those categories AFAIK. Don't know about now, but the old TwinMOS BH-5 was nothing to write home about.

[enok]

good morning BH-5! :toast: a new day as araise for u babe!

guess i will have to sell my UTT afterall... :D

[Nazaar]

Are we talking about new ram modules with the "classic" bh-5 chips on ?

[P_1]

are u guys pumping out new chips or using old stock?

[BeSaiD]

I'll looking forward to their arrive to Spain! :D

[Zeus]

Bring on the BH-5 for €39,- a stick!! :banana:

[Entity_Razer]

I just hope such things don't elevate Twinmoss into the "performace"class producersd

because as long as I know twinmoss always had decent ram for cheap prices.

I mean I won't have a problem if they up prices by say 10 dollars or something, ok this is to cover their expenses but I'd hate to see twinmoss turn into a OCZ like bussness (or corsair) forwhich you have to lay down 200euros to get some decent ram :banana: :banana: :banana: :banana: :banana:

[TEDY]

i second that entity.....it's fine few % margain ...

[WiCKeD]

It'd be interesting to know! While TCCD can do high FSB, BH5 can do comparatively lower frequencies with tight timings! :D

Why did Winbond exit the RAM market anyway?
http://forums.extremeoverclocking.com/showthread.php?p=990327

[Holst]

What are you talking about, TwinMOS is foolish.

Not all BH-5 is good BH-5. In fact, there is still a lot of BH-5 chips out there on the market. However, most of the "overclocking" weeks have already been bought up, and the remaining chips are the kind that people get and whine "buhuhuhoo my BH-5 doesn't go over 220MHz halp meh!", and then wonder why. Of course it's a combination of things - sh1tty PCB, bad week, lack of any binning out of bad chips... cheap materials. And TwinMOS fits into all of those categories AFAIK. Don't know about now, but the old TwinMOS BH-5 was nothing to write home about.

I dont agree with what you say about twinmos poor quality.

I have personally used many of there Bh5 sticks, and still own 4 lots of 256mb.

All 4 sticks are pretty evenly matched (topping out around 260-265mhz tight timings and 3.4v)
They have all survived 18months plus of 3.2v minimum voltage on socketA and A64.

Admittedly the twister series and some of there other products fall short of the mark, but there "budget" bh5 has been superb for me, and many others.

And it is cheap ram, my sticks all cost me £25 each!!!, 50%+ less than OCZ and corsair, and my sticks clock as good, if not better.

Obviously you dont get any guarantees with twinmos, but its so cheap you can resell it for almost what you pay if its not as good as you hoped.

[BeSaiD]

I'm with entity, too. I don't like to pay so much for memory... :(

[tictac]

because as long as I know twinmoss always had decent ram for cheap prices.


Agreed. :toast:

[TEDY]

of topic : tictac how was exam ? :D

[esdee]

alonso correct me if im wrong but i remember at hipro5's lab tha he did 270 with 2x512 twinmos BH5 a few weeks ago!

[felinusz]

What are you talking about, TwinMOS is foolish.

Not all BH-5 is good BH-5. In fact, there is still a lot of BH-5 chips out there on the market. However, most of the "overclocking" weeks have already been bought up, and the remaining chips are the kind that people get and whine "buhuhuhoo my BH-5 doesn't go over 220MHz halp meh!", and then wonder why. Of course it's a combination of things - sh1tty PCB, bad week, lack of any binning out of bad chips... cheap materials. And TwinMOS fits into all of those categories AFAIK. Don't know about now, but the old TwinMOS BH-5 was nothing to write home about.


Most BH-5 doesn't perform (waah, my BH-5 won't clear 220, as you say) untill ~3V. Most people do not have 3V of VDIMM available on their boards without modifications. BH-5 will not usually clear 220 with only ~2.85V. As for "overclocking steppings"; There are good and bad examples of almost all BH-5 steppings - some chips are weak, and some overclock really well.

I agree though, Twinmos RAM generally borders on generic quality; the PCB and speed bin make *ALL* the difference when it comes to getting good overclocking RAM.

[Entity_Razer]

I defenitly don't support iddqd's view.

I never, ever had problems with twinmoss. As said before me it is off course true that with twinmoss you don't get a guarantee that it'll reach xxxmhz or something but then again it IS valueram now isn't it?

and also iddqd, I don't feel that you are in any right of complaining seeing as Twnmos now uses BRAINPOWER pcb's on their VALUEram....

I don't know about you but imho that shows a devotion to their costumers and work and the fact that they do not belong in the craphouse....

They deliver decent memory and for very good prices. IF you want assured overclocks of 250MHZ @ 5-2-2-2 then you should go OCZ VX or something and pay 300 euros or more for a gig.

for me twinmoss will suffice nicely and I think I found a constant in my life at this moment. I always used to take Apacer but after my good twinmoss experience (no not UTT, Mtecs clocking 240@ 7-3-3-2T @ 2.9Volts)
and a good experience with a frend of mines year old RAM which also contains brainpower PCB's I can say nothing evil/bad about Twinmoss except that they might wanna put their valueram in a package also. I wouldn't mind paying 2 euros more fore a package bevause at this moment I don't like the bare modules when I take'm home, I like'm wrapped to protect them nicly.

Thats the only thing I can think of to critisice Twinmoss with.

[Rabbi_NZ]

Alonso, when you get the first shipment would you post the "Week" of the chips used please?

[enzoR]

so winbond is makeing BH-5 again huh? :confused:

iddq afaik almost all BH5 will do 250 with 3.5v max. :stick:

[TEDY]

enzor : mine hit 250 with 3.3...only if i had 3.4-3.5 :)

[perry_78]

Ah, lack of capitol. If those are indeed the secrets behind Windbond's disappearance, then i doubt that they would kick in long after their exit.

If anyone has any links or information regarding the disappearance of winbond from the RAM market, pleas PM me. I would like to gather as much information on the topic.

Hey, for £25 a stick, it's FREE! :D

[jjcom]

Saaya did something on that I believe one, try PMing him

jjcom

[Garrett]

What are you talking about, TwinMOS is foolish.

...but the old TwinMOS BH-5 was nothing to write home about.
Nah... my twinmos BH-5 sucks too *cough*... "only" DDR540 2-2-2-5 @ 3.6v ;)

That's Dual Channel (socket 939) of course... not just 1 stick :stick:

I managed to get some 44D AA4T sticks (2x512) and I'll pick 'em up this week... see what they can do ;)

But I do like my BH-5's a lot... selling those will be hard to me ;)

[iddqd]

Most BH-5 doesn't perform (waah, my BH-5 won't clear 220, as you say) untill ~3V. Most people do not have 3V of VDIMM available on their boards without modifications. BH-5 will not usually clear 220 with only ~2.85V. As for "overclocking steppings"; There are good and bad examples of almost all BH-5 steppings - some chips are weak, and some overclock really well.

I mean whatever voltage you throw at it. Some BH-5 will start gaining 1Mhz every .1vdimm after 220MHz... go figure. Oh wait, Buffalo! ;)

so winbond is makeing BH-5 again huh? :confused:

iddq afaik almost all BH5 will do 250 with 3.5v max. :stick:
Winbond is dead. :stick:. Whatever Infineon is making on their fab, is their concern. Most likely a tweaked, or a smaller process CH-5 (CH-4?, in this case?) Winbond UTT is just plain WTF:confused:. You guys made it up, based solely that the packaging appears to be the same as Winbond's (two dimples). However, the packaging machine that makes those dimples is very common in Taiwan. For example, you wouldn't be able to tell the difference between unmarked ProMOS and Winbond chips.

Nah... my twinmos BH-5 sucks too *cough*... "only" DDR540 2-2-2-5 @ 3.6v ;)

That's Dual Channel (socket 939) of course... not just 1 stick :stick:

I managed to get some 44D AA4T sticks (2x512) and I'll pick 'em up this week... see what they can do ;)

But I do like my BH-5's a lot... selling those will be hard to me ;)
My OCZ BH-5 does 250MHz 2-2-2 at only 3.1vdimm. 3.35, and I got 270 :p:. Also, I'm under impression that you are using 2x256 sticks to achieve 270MHz. I'm using 2x512 sticks. If you don't beleive me, check the BH-5 database we have stuck to the top, TwinMOS section does not impress :yawn:.

[Rabbi_NZ]

Winbond is dead. :stick:. Whatever Infineon is making on their fab, is their concern. Most likely a tweaked, or a smaller process CH-5 (CH-4?, in this case?) Winbond UTT is just plain WTF:confused:. You guys made it up, based solely that the packaging appears to be the same as Winbond's (two dimples). However, the packaging machine that makes those dimples is very common in Taiwan. For example, you wouldn't be able to tell the difference between unmarked ProMOS and Winbond chips.
no one made anything up, there are news articles about Winbond making chips again and selling them to Infineon, and Alonso himself just said UTT is basically CH5... which we already figured for ourselves.

CH just means the size of he process, and 5 jus means 5ns... so when Twinmos sells 50D UTT chips they ARE CH5.

[iddqd]

CH is just a revision code. There is also AH, BH, DH, GH. What they mean is a mystery to me.

[Rabbi_NZ]

CH is just a revision code. There is also AH, BH, DH, GH. What they mean is a mystery to me.
you said it right there dude.

to be honest I have not heard of DH or GH but maybe they exist.

CH means 130nm process... CH6 & CH5 are the same process, same chips, from the same machines, possibly same date, same size, same mah and pah.
That's why Alonso said they were CH5 (in first post), because they are based on the same process... all be it improved by the looks

[tictac]

What are you talking about, TwinMOS is foolish.

Not all BH-5 is good BH-5. In fact, there is still a lot of BH-5 chips out there on the market. However, most of the "overclocking" weeks have already been bought up, and the remaining chips are the kind that people get and whine "buhuhuhoo my BH-5 doesn't go over 220MHz halp meh!", and then wonder why. Of course it's a combination of things - sh1tty PCB, bad week, lack of any binning out of bad chips... cheap materials. And TwinMOS fits into all of those categories AFAIK. Don't know about now, but the old TwinMOS BH-5 was nothing to write home about.

you deserve a :slap: in the face

[kristos]

so are these BH-5's new chips or old stock?


and, not that I expect this one answered (not yet anyway) but, how do I tell them apart from the rest :cool:

[metro.cl]

are these like the same bh5??? or is it an improoved revision that would do 300mhz 2-2-2-11 @ 3.3v :)

p.d. hope they are a new revision like i say

[Craig]

BUT:

today i have been talking to our technical engineers and also took a close look on the price list and what did i see?

Speed Premium DDR450 with BH-5

Hope to get some samples asap to tell ya the results ;)


Need more info on this please.

And for those saying BH5 RAM isn't all that great........I'm typing along with BH5 @ 227 X 11.5 and tight timings 2-2-2-7-10-13........and 2.9vdimm :D

I'll grant that I'm not making use of Winbond sticks, but they sure are Winbond chips. Mine are Mushkin Lvl II black PC3500 sticks. They'll do 238 at the above timings with 3.3vdimm. I've 6 other sticks here, and the least of them does 227 dual channel at tight timings.

I've used Kingston, OCZ & Mushkin.....all have good stuff & so did Winbond.

I can't hardly wait for the NF4 from DFI with it's 4 volt vdimm ......I"m going to build a A64 and put in a gig and a half of my BH5 Mushkin at tight timings. She'll fly......:D

But if Winbond really is about to release DDR450 BH5 sticks I'll be buying a gig or more for sure. If rated for 450 they'll do even better with some juice to them......:YIPPIE: I'd love to have a gig @ DDR500 & tight timings. And I also hope they sell the sticks with the chips bare just as they did before.

Best news I've read in a good while regarding RAM! :banana:

[[XC] serlv]

slightly OT...but


UTT

I have seen these three letters a lot in the last 2 weeks, and obviously I know it relates to RAM, or some form of it. But that's all I know.
Somebody clue me in, please.
Been bugging the crap out of me, every time I see it.

Scott

[reject]

we think it means UnTesTed
they just pumping out chips and selling them striaght away, to vendors who like ocz test them and slap a hefty price tag on
or twinmos who let us test them ;)

[bachus_anonym]

we think it means UnTesTed
they just pumping out chips and selling them striaght away, to vendors who like ocz test them and slap a hefty price tag on
or twinmos who let us test them ;)
:lol: that just made my crappy day so far :lol:

[craig588]

I thought they didn't have a real name so people made up something untill we saw the unrebadged version. The UTT stood for ultra tight timings. Guess I was wrong.

[sjohnson]

Bringing up Buffalo sticks, anyone with NF3/NF4 have a pair to test in an A64 environment. The Buffalo I bought ran great in my kd7-e (220+ FSB @ 3.1v) and my DFI infinity (254-255 @ 3.3v) with tight timings, so I wonder if 260-280 isn't possible with Buffalo on an A64 rig. I'd do it if anyone wants to "loan" a 64 rig and processor :D

[iddqd]

you deserve a :slap: in the face
Trust me, you don't want me to move to personal insults. Cause then you'll be wearing three boots, two on your feet and one *****.

you said it right there dude.

to be honest I have not heard of DH or GH but maybe they exist.

CH means 130nm process... CH6 & CH5 are the same process, same chips, from the same machines, possibly same date, same size, same mah and pah.
That's why Alonso said they were CH5 (in first post), because they are based on the same process... all be it improved by the looks
Clearly, that's not the case, as BH-6 is 166nm gate length (or as you like to call it, "process") and BH-5 is 150nm. BH-4 was supposed to be 130nm.


// edited for flaming.

[Super strokey]

hahahaha, mercy! I think your not sure who your talking to lol

[EnJoY]

Arguing on the internet is like the special olympics. Even if you win, you're still retarded.

[Super strokey]

Arguing on the internet is like the special olympics. Even if you win, you're still retarded.


lol, thats awesome! Mind if i sig it?

[sjohnson]

lol, thats awesome! Mind if i sig it?Here's the pic it came from

[EnJoY]

Yea sig it and say it's my quote! :D

[Bluebeard]

*UPDATE*

BH-5 will come back in old technologie as usual in 0,175um
CH-5 are, as everybody knows, in new technologie 0,13um
DH-5 will hopefully be here soon as well in 0,11um

BH-5 will be used for DDR450 Speed Premium so there should be no "week-hunting" ;)

[CTKP]

so will the new speed premium's have bh-5 chips from old stock, or will they be from a new production?

[Bluebeard]

as long as there was no Winbond production for long, long time, these will be new produced ones ;)

DDR400 Speed Premium to come witn CH-5
DDR450 Speed Premium (not released yet) to come in BH-5

[Garrett]

...My OCZ BH-5 does 250MHz 2-2-2 at only 3.1vdimm. 3.35, and I got 270 :p:. Also, I'm under impression that you are using 2x256 sticks to achieve 270MHz... :yawn:.
Why don't you kick 3.6v through it then? Might get you 290 ;)
Since we're busy why don't you make that 3.8v... you could actually get 300 ;)

[Nazaar]

omg thats amazing !! When will they hit the market ?

[Bluebeard]

soon, i guess ;)

first samples to arrive next week... after testing the party can start :D

[El Snorro]

What are you talking about, TwinMOS is foolish.

Not all BH-5 is good BH-5. In fact, there is still a lot of BH-5 chips out there on the market. However, most of the "overclocking" weeks have already been bought up, and the remaining chips are the kind that people get and whine "buhuhuhoo my BH-5 doesn't go over 220MHz halp meh!", and then wonder why. Of course it's a combination of things - sh1tty PCB, bad week, lack of any binning out of bad chips... cheap materials. And TwinMOS fits into all of those categories AFAIK. Don't know about now, but the old TwinMOS BH-5 was nothing to write home about.

Nothing to write homa about???
Why not?
My 2x512 twinmos bh5 (week 13) does 275mhz 2t and 255 mhz 1t (benchable) @3,6v which in still in the safe zone according to winbond.
But I have sold this bh5 to buy some OCZ VX (higher speeds @1T)

Will this new bh5 chips do any better than 'old' bh5??
And will the high speeds= 2T problem will be solved??
Because that will be some points for me to make my decision to wait for this SP or to buy some OCZ VX...

[BeSaiD]

DH-5 in Speed Premium DDR500?? :D

[Bluebeard]

released as DDR450 2,5-3-3-8 i´m sure these modules will rock for standard users as well as for OC

TwinMOS will never special select like OCZ, G.Skill or others out there - but is it really worth it? Look at the UTTs...

TCCD - fast timings/ high fsb with low voltage
CH-5 - love to get V
BH-5 - fast timings with low voltage / high fsb has to be testes -> i think there still can be differences even the process and production is the same - we have to see

both CH-5 and BH-5 will be much cheaper than TCCD ;)

[Garrett]

DH-5 in Speed Premium DDR500?? :DI'm very curious
http://www.frags4fun.com/modules/Forums/images/smiles/carrot.gifhttp://www.frags4fun.com/modules/Forums/images/smiles/carrot.gifhttp://www.frags4fun.com/modules/Forums/images/smiles/carrot.gifhttp://www.frags4fun.com/modules/Forums/images/smiles/carrot.gif

[enzoR]

enzor : mine hit 250 with 3.3...only if i had 3.4-3.5 :)

yea i'm saying aslong as the BH5 isnt defective all of it will hit 250 with a maximun voltage required of 3.5. some will do so much better with 3.5

[enzoR]

Trust me, you don't want me to move to personal insults. Cause then you'll be wearing three boots, two on your feet and one in your ass.

Clearly, that's not the case, as BH-6 is 166nm gate length (or as you like to call it, "process") and BH-5 is 150nm. BH-4 was supposed to be 130nm.


sorry but that info is wrong. The XH means die chip, 5/6/7.5 is speed grade. all CH come from same die. same with BH and AH. all the same manufacturing process. these dies get speed binned and they get their ns rateings handed to them. call winbond up for yourself, or send them an email. ;)

[tictac]

both CH-5 and BH-5 will be much cheaper than TCCD ;)

Cool.... :toast:

[esdee]

*UPDATE*
DH-5 will hopefully be here soon as well in 0,11um

BH-5 will be used for DDR450 Speed Premium so there should be no "week-hunting" ;)

ok ... DH-5 soon ... BH-5 on speed-premium
CH-5 and UTT currently on them, i can see no reason why not buying speed premioun !!!

i just CAN"T Wait!

[CrimeDog]

DDR400 Speed Premium to come witn CH-5
DDR450 Speed Premium (not released yet) to come in BH-5
So is that 100% winbond chips or a mixed bag like the speed premium with the UTT chips?
Pretty crazy Winbond getting back into the ddr market, it's gotta be a temperary thing or the ddr2 market totally bombed.

edit: i just got what you meant by week hunting, d'oh :toast:

[Bluebeard]

your comment makes no sense to me.... sry

http://www.digitimes.com/bits_chips/a20050125A4015.html

[gundamit]

http://www.digitimes.com/bits_chips/a20050125A4015.html
Holy ... ! $8,000,000,000 loan? :eek: Well I'll try to do my part helping them pay back their loan, but I can only kick in $200 tops for a gig of the new/old DDR they'll be producing. Anyone else want to cover the rest? :rolleyes:

[Nazaar]

whats this DH-5 ? Better then utt/bh-5 ??

[DaWaN]

*UPDATE*

BH-5 will come back in old technologie as usual in 0,175um
CH-5 are, as everybody knows, in new technologie 0,13um
DH-5 will hopefully be here soon as well in 0,11um

BH-5 will be used for DDR450 Speed Premium so there should be no "week-hunting" ;)

Will there be also some 4ns stuff? Bh-4 or DH-4, that would be great :toast:
I'm also very interrested in DH-5, should it be better than BH-5?

[esdee]

your comment makes no sense to me.... sry

http://www.digitimes.com/bits_chips/a20050125A4015.html

holly :banana:

[bobmanfoo]

.....
both CH-5 and BH-5 will be much cheaper than TCCD ;)
k, im a bit confused now, ch-5= utt -->TM ddr400 speed premium? how much will be bh-5 sticks be compared to the TMSP400? 160$ for a gig of the tmsp400 here in the states

[seldomsean]

holly :banana:


Well i say holy :banana4:

[Entity_Razer]

BH 5 comming back? .......


:slobber: :slobber: :slobber:

Although never owning them I've heard soooo much about them.

Can't wait to pump (yet again) some money in some sticks to test that RAM

BTW on which sticks will these be?

Speed premium I'm guessing?

BTW can you say a directional price?

for 512 MB,.... 80$,70$? :p

Thanks for the good news alonso!!!!

[WiCKeD]

I'm also very interrested in DH-5, should it be better than BH-5?DH-6 is the crap on the bottom of my Seagate SCSI drive. No joke.

Oh and...


http://www.digitimes.com/bits_chips/a20050125A4015.htmlHOLY :banana::banana::banana::banana:!!!

[enzoR]

but DH isnt even being made yet?...

[WiCKeD]

...then how did it get on the bottom of my drive? It's made. They used it to designate another type of IC.

wait a minute... who cares?! Did you even read that link? Winbond is back! :bounce:

[enok]

crossing my fingers for a possible BH-4... :D

[sapphire]

Wel, nice to know that BH-5 is comming back but stil I'm not throwing a party just yet.
Imho BH-5 isn't the super OC ram, everyone is talking about, has anyone ever seen 512 BH-5 stick do 270 2-2-2-X 1t ? Personally I haven't.

So I would be great is the new batch of BH-5/DH-5 stick would have better preformance ecspecially @ 1t for al the A64 user out there :)

[alexio]

I think the Brainpower pcb makes the 512mb bh-sticks scale allot higher, just as it did with the ch-5 512- and 256mb-sticks.

[Zeus]

Wel, nice to know that BH-5 is comming back but stil I'm not throwing a party just yet.
Imho BH-5 isn't the super OC ram, everyone is talking about, has anyone ever seen 512 BH-5 stick do 270 2-2-2-X 1t ? Personally I haven't.

So I would be great is the new batch of BH-5/DH-5 stick would have better preformance ecspecially @ 1t for al the A64 user out there :)

You must be confused my friend, not being able to go to xtreme hights with 2x512mb is not all about the RAM, the memorycontroller comes into play big time here.

Seeing Opp's 43K score i have to admit there's a good point in using 2x512mb though. :D

If you are heading towards VX being the best OC mem i think you're heading the wrong direction, at least, if VX or UTT is CH-5.(which i assume it is)

It is a proven fact that BH-5 performs better than CH-5, same speed and settings.

I'm afraid BH-5 will still hold the crown. ;)

[seldomsean]

You must be confused my friend, not being able to go to xtreme hights with 2x512mb is not all about the RAM, the memorycontroller comes into play big time here.

Seeing Opp's 43K score i have to admit there's a good point in using 2x512mb though. :D

If you are heading towards VX being the best OC mem i think you're heading the wrong direction, at least, if VX or UTT is CH-5.(which i assume it is)

It is a proven fact that BH-5 performs better than CH-5, same speed and settings.

I'm afraid BH-5 will still hold the crown. ;)


I wonder if winbond had this in their plans all along...

Leave the market and them make a Comeback !!

Kinda makes sense in maketing terms doesnt it :confused:

[Bluebeard]

for DH-5 we have to wait longer, while others (e.g PSC) already have their process down to 0,11um

@Zeus

;)

i´d really love to see them performing on a nice 6-layer PCB with hipro´s maximizer sittin´in the back of a nice MSI Neo ;)

[Rabbi_NZ]

Trust me, you don't want me to move to personal insults. Cause then you'll be wearing three boots, two on your feet and one *****.

Clearly, that's not the case, as BH-6 is 166nm gate length (or as you like to call it, "process") and BH-5 is 150nm. BH-4 was supposed to be 130nm.


// edited for flaming.

sorry but that info is wrong. The XH means die chip, 5/6/7.5 is speed grade. all CH come from same die. same with BH and AH. all the same manufacturing process. these dies get speed binned and they get their ns rateings handed to them. call winbond up for yourself, or send them an email. ;)
I, and most people round here, will agree with ensoR's statement.

iddqd,
if you dont have something intelligent (or at least nice) to say....

[esdee]

i´d really love to see them performing on a nice 6-layer PCB with hipro´s maximizer sittin´in the back of a nice MSI Neo ;)

heh the description matches my rig too ;)

[enzoR]

Lucky B$#@%*^

[sapphire]

You must be confused my friend, not being able to go to xtreme hights with 2x512mb is not all about the RAM, the memorycontroller comes into play big time here.

Seeing Opp's 43K score i have to admit there's a good point in using 2x512mb though. :D

If you are heading towards VX being the best OC mem i think you're heading the wrong direction, at least, if VX or UTT is CH-5.(which i assume it is)

It is a proven fact that BH-5 performs better than CH-5, same speed and settings.

I'm afraid BH-5 will still hold the crown. ;)
That I know, i have some experience with both BH-5 and CH-5 ;)
And also i'm not saying that VX is the best OC mem.

But your saying the memory controller comes into play here, the only experience I have with winbond mem is on nForce2 bords, you think the A64 on-die mem controller handles 2x512mb at high speeds better than the 'old' memory controllers? Cause that would be great :D

[Rabbi_NZ]

...you think the A64 on-die mem controller handles 2x512mb at high speeds better than the 'old' memory controllers? Cause that would be great :D
No doubt about it, it DOES :toast:

[P_1]

jesus this is great news. I wonder how much the memory prices are gonna drop with the advent of winbond being back. :slobber:

[Bennah]

DDR450 Speed Premium (not released yet) to come in BH-5

3700 BH-5, get me down for some :D :D

[jjcom]

Man, Winbond is back...YES! I wasn't in the overclocking thing when Winbond was going. Now maybe I can get some BH-5 on some of those brainpower PCB...that'd be nice

jjcom

[Rabbi_NZ]

3700 BH-5, get me down for some :D :D
PC3600 ;)

[Super strokey]

Man, Winbond is back...YES! I wasn't in the overclocking thing when Winbond was going. Now maybe I can get some BH-5 on some of those brainpower PCB...that'd be nice

jjcom

that would make for soem impressive dimms indeed,

[jjcom]

Well I plan to upgrade this summer or maybe early '06 and by then Winbond should hopefully be making BH-5 and that in large enough amounts that I could get ahold of some :D

jjcom

[situman]

I might be wrong, but the smaller the process, the less volts it can sustain am I correct? Also, the new winbond chips will be based on INFINEON tech (based on the article), since when did anyone overclock with INFINEON?? If the new winbond chips shrink their process AND uses Infineon tech, I think the day of super clocks with the tightes timings will be over. Btw someone told me BH-5 doesnt like anything above Cas 2, not sure if its true. Looking at the new BH-5 at DDR450, its timings might suggest tweaked BH-5.

[trans am]

is anyone concerned that AMD is going to use DDRII in 2006? I mean this is great news I guess, but how long is it going to last? Unless that old winbond machine can make DDRII?

[jjcom]

I would guess that it will. Wouldn't make much since for Winbond to make a huge loan like that and make only a type of RAM that won't be around long enough to pay back the loan.

Thats the RATED timings. BH-5 doesn't like CAS3 and it'll work with CAS2.5. TwinMOS makes it rated at different timings than what it can do. The current TwinMOS Speed Preuium chips are rated at CAS 2.5 @ DDR400 as well

jjcom

[trans am]

winbond ddr2? I would love to see that. I hope it works out. :)

[STEvil]

Bringing up Buffalo sticks, anyone with NF3/NF4 have a pair to test in an A64 environment. The Buffalo I bought ran great in my kd7-e (220+ FSB @ 3.1v) and my DFI infinity (254-255 @ 3.3v) with tight timings, so I wonder if 260-280 isn't possible with Buffalo on an A64 rig. I'd do it if anyone wants to "loan" a 64 rig and processor :D

Got my 2x512 Buffalo CH-5 to 245 2-2-2-x on my 754 2800+ with 3.30v

[MooCow2]

Holy ... ! $8,000,000,000 loan? :eek: Well I'll try to do my part helping them pay back their loan, but I can only kick in $200 tops for a gig of the new/old DDR they'll be producing. Anyone else want to cover the rest? :rolleyes:

NT dollars = New Taiwan dollars

32NT = 1USD

[Susquehannock]

Interesting coincidence there is new Bh-5 coming out now that the new TCCD
are doing so well. ;) :D

[enok]

considering the loss of performace going from BH-5 to CH-5 (180nm to 130nm), i think DH-5 (110nm) won't give any advantage. :(

[Nazaar]

man, cant wait for these modules to hit the stores in belgium !!

[enzoR]

are the new twinmoss based on BH5 being mass manufacutred?

[iddqd]

whats this DH-5 ? Better then utt/bh-5 ??
No idea. I've seen it on a hard drive. (So, I'm guessing it's not that great)

I would guess that it will. Wouldn't make much since for Winbond to make a huge loan like that and make only a type of RAM that won't be around long enough to pay back the loan.

Thats the RATED timings. BH-5 doesn't like CAS3 and it'll work with CAS2.5. TwinMOS makes it rated at different timings than what it can do. The current TwinMOS Speed Preuium chips are rated at CAS 2.5 @ DDR400 as well

jjcom
So, Winbond is back in action? Most interesting.

Nothing to write homa about???
Why not?
My 2x512 twinmos bh5 (week 13) does 275mhz 2t and 255 mhz 1t (benchable) @3,6v which in still in the safe zone according to winbond.
But I have sold this bh5 to buy some OCZ VX (higher speeds @1T)

Will this new bh5 chips do any better than 'old' bh5??
And will the high speeds= 2T problem will be solved??
Because that will be some points for me to make my decision to wait for this SP or to buy some OCZ VX...
Funny, how my BH-6 can do the same thing. And they're 512 sticks. Edit: just noticed that was 2t. The BH-6 does 275MHz 2-2-2 1t

Why don't you kick 3.6v through it then? Might get you 290 ;)
Since we're busy why don't you make that 3.8v... you could actually get 300 ;)
I would, but this stupid PSU won't go higher than 3.5v on the +3.3 line :(. I need to get a new PSU.. A PCP&C510D or something.

[EMU]

Do any of you guys have an idea what chips are used on the Twinmos TMII400? Are these UTT as well? If not any comparison of performance?

Really glad Winbond is back in the market :D

[situman]

looks like Ebay BH-5s will tank.

[Mag Master 21]

Wel, nice to know that BH-5 is comming back but stil I'm not throwing a party just yet.
Imho BH-5 isn't the super OC ram, everyone is talking about, has anyone ever seen 512 BH-5 stick do 270 2-2-2-X 1t ? Personally I haven't.

So I would be great is the new batch of BH-5/DH-5 stick would have better preformance ecspecially @ 1t for al the A64 user out there :)

Believe it.. This is a 512 stick at 270 with 2-2-2-10 fully A64 Tweaked...

http://server3.uploadit.org/files/magmaster21-270fsb.JPG

[enzoR]

damn man thats insane! what volts?

[Nube]

Believe it.. This is a 512 stick at 270 with 2-2-2-10 fully A64 Tweaked...

http://server3.uploadit.org/files/magmaster21-270fsb.JPG
I think he is talking about 2*512 dual channel... try get 270 2-2-2 then...

[enzoR]

if the ram can do it in single, the ram can do it in dual. it all depends on the memory controller if it can still do it in dual.

[Mag Master 21]

damn man thats insane! what volts?

that's with 3.6vdimm from a 3.72 rail.. I'm hoping to get a booster soon so I can really push her ;)

[Rabbi_NZ]

BH6/BH5 not doing well in 2x512 is a myth in my opinion... dont know where it started but Ive seen PLENTY of high 2x512 BH6/BH5 clocks

[Revv23]

2*512 all depends on the memory controller...

sounds like AMD will be making controller upgrades soon so this all sounds good to me... :)


BTW i disagree with what you guys are saying about the processes...

http://www.xtremesystems.org/forums/showthread.php?t=43707&highlight=BH-5+legend


AH-6 180nm
BH-6 166nm
CH-6 166nm
AH-5 (doesnt exist?)
BH-5 150nm
CH-5 130nm
BH-4 130nm[/CENTER]

this explains why bh-5 and bh-6 perform almost the same, they are made in almost the same manufacturing process.



according to sayaa BH-5 is 150nm, bh-6 is 166nm, which is why BH-6 can handle REALLY high volts better then BH-5, just ask hipro5 about that one...also, it is my understanding that CH-5 is not the same design as BH-5, as CH-5 was made to be cheaper, explaining why it performed worse then BH at same clocks/timings....

[Rabbi_NZ]

Revv23,

bigtoe himself has said many times BH5=BH6 & CH5=CH6, same die jus different speed bins. Until I see where saaya got his info Im sticking with what bigtoe said.

[enzoR]

ive seen a german site post that info.

[Susquehannock]

3.6v ... thought so. The TCCD should be able to match that at
lower voltages right?

[HARDCORECLOCKER]

:D THIS is insane:

http://de.geocities.com/hardcoreclocker/Corsair275.JPG

"OLD" Corsair BH-5 :toast:

[hipro5]

I think he is talking about 2*512 dual channel... try get 270 2-2-2 then...

This was done with Dual channel BH-5 2 x 512MB Twinmos sticks - old school ones - PC3200 at 3.62Volts......Thanks Alex...... ;) :D

http://www.thelab.gr/images/Hipro5/P4_3.8GHz/png/Super-Pi_1M_unpatched_25sec_1_1_5110MHz_png.png

[SlackeR]

Maximizer sure seems nice ;)
Very impressive clocks, Hipro!

[Revv23]

Revv23,

bigtoe himself has said many times BH5=BH6 & CH5=CH6, same die jus different speed bins. Until I see where saaya got his info Im sticking with what bigtoe said.


i stand corrected.

[hipro5]

Maximizer sure seems nice ;)
Very impressive clocks, Hipro!

It sure does........Thanks M8...... :D

[clarence]

i stand corrected.

No need to argue, just read up on their respective datasheets from Winbond.

From the Winbond 942516CH datasheet (for all speed grades from CH-75 to CH-5), the manufacturing process is described as 0.13um. It also says the 942516CH is sorted into 4 speed grades (-75,-7,-6,-5).

I would be very surprised if Winbond has lines that produce ram using 0.18um, 0.166um, 0.15um & 0.13um in an overlapped timeframe. For starters, it means massive investment wasted, as 0.166 & 0.15 are not necessary as Winbond is already migrating to 0.13 at the time when the BH series was in production. The incremental benefits offered by 0.166 & 0.15 over 0.18 does not justify the investment needed. Secondly, why would Winbond waste production cost in producing chips in a larger process when a smaller & more cost effective process is available (not to mention the cost in inventory controls)? Thirdly, what happens to those CH-5 that cannot function at 200MHz but can do so at 166MHz?? Do they junk them altogether?? They surely can't be called CH-6 if it's from a different process!!!

The words AH, BH, CH tells u which generaation it belongs, & the -75, -7, -6, -5 tells u it's speed. If u look at the ICs themselves, it's printed "BH-5" or "CH-5", BH5 or CH5 is just an abbreviation.

http://www.winbond.com.tw/c-winbondhtm/partner/b_2_g_1_a.htm#06

Datasheet for CH series available at bottom of the page.

[jumanji969]

3.6v ... thought so. The TCCD should be able to match that at
lower voltages right?

Yea with 2.5-3-3-7 timings not those nice 2-2-2-x timings.

[Rabbi_NZ]

No need to argue, just read up on their respective datasheets from Winbond.

From the Winbond 942516CH datasheet (for all speed grades from CH-75 to CH-5), the manufacturing process is described as 0.13um. It also says the 942516CH is sorted into 4 speed grades (-75,-7,-6,-5).

I would be very surprised if Winbond has lines that produce ram using 0.18um, 0.166um, 0.15um & 0.13um in an overlapped timeframe. For starters, it means massive investment wasted, as 0.166 & 0.15 are not necessary as Winbond is already migrating to 0.13 at the time when the BH series was in production. The incremental benefits offered by 0.166 & 0.15 over 0.18 does not justify the investment needed. Secondly, why would Winbond waste production cost in producing chips in a larger process when a smaller & more cost effective process is available (not to mention the cost in inventory controls)? Thirdly, what happens to those CH-5 that cannot function at 200MHz but can do so at 166MHz?? Do they junk them altogether?? They surely can't be called CH-6 if it's from a different process!!!

The words AH, BH, CH tells u which generaation it belongs, & the -75, -7, -6, -5 tells u it's speed. If u look at the ICs themselves, it's printed "BH-5" or "CH-5", BH5 or CH5 is just an abbreviation.

http://www.winbond.com.tw/c-winbondhtm/partner/b_2_g_1_a.htm#06

Datasheet for CH series available at bottom of the page.

Thank you clarence

[wespwnzu]

im still using my khx3500 bh-5 and loving it, except for the fact that it wont go over 3200 speeds, and i cant figure out why. i think its just this :banana::banana::banana::banana:ty mobo, i mean i got them from oppainter so it cant be the sticks

[J-Pak]

Sounds very promising :)

[enzoR]

No need to argue, just read up on their respective datasheets from Winbond.

From the Winbond 942516CH datasheet (for all speed grades from CH-75 to CH-5), the manufacturing process is described as 0.13um. It also says the 942516CH is sorted into 4 speed grades (-75,-7,-6,-5).

I would be very surprised if Winbond has lines that produce ram using 0.18um, 0.166um, 0.15um & 0.13um in an overlapped timeframe. For starters, it means massive investment wasted, as 0.166 & 0.15 are not necessary as Winbond is already migrating to 0.13 at the time when the BH series was in production. The incremental benefits offered by 0.166 & 0.15 over 0.18 does not justify the investment needed. Secondly, why would Winbond waste production cost in producing chips in a larger process when a smaller & more cost effective process is available (not to mention the cost in inventory controls)? Thirdly, what happens to those CH-5 that cannot function at 200MHz but can do so at 166MHz?? Do they junk them altogether?? They surely can't be called CH-6 if it's from a different process!!!

The words AH, BH, CH tells u which generaation it belongs, & the -75, -7, -6, -5 tells u it's speed. If u look at the ICs themselves, it's printed "BH-5" or "CH-5", BH5 or CH5 is just an abbreviation.

http://www.winbond.com.tw/c-winbondhtm/partner/b_2_g_1_a.htm#06

Datasheet for CH series available at bottom of the page.


Myth Busted :D
someone should inform Saaya to correct his BH database.

[P_1]

No need to argue, just read up on their respective datasheets from Winbond.

From the Winbond 942516CH datasheet (for all speed grades from CH-75 to CH-5), the manufacturing process is described as 0.13um. It also says the 942516CH is sorted into 4 speed grades (-75,-7,-6,-5).

I would be very surprised if Winbond has lines that produce ram using 0.18um, 0.166um, 0.15um & 0.13um in an overlapped timeframe. For starters, it means massive investment wasted, as 0.166 & 0.15 are not necessary as Winbond is already migrating to 0.13 at the time when the BH series was in production. The incremental benefits offered by 0.166 & 0.15 over 0.18 does not justify the investment needed. Secondly, why would Winbond waste production cost in producing chips in a larger process when a smaller & more cost effective process is available (not to mention the cost in inventory controls)? Thirdly, what happens to those CH-5 that cannot function at 200MHz but can do so at 166MHz?? Do they junk them altogether?? They surely can't be called CH-6 if it's from a different process!!!

The words AH, BH, CH tells u which generaation it belongs, & the -75, -7, -6, -5 tells u it's speed. If u look at the ICs themselves, it's printed "BH-5" or "CH-5", BH5 or CH5 is just an abbreviation.

http://www.winbond.com.tw/c-winbondhtm/partner/b_2_g_1_a.htm#06

Datasheet for CH series available at bottom of the page.

According to that wouldnt that also mean that theres no reason for winbond to be producing bh5 and ch5, since they have a new proccess and it would be a waste if they started producing old chips on an old proccess. :stick:

[clarence]

According to that wouldnt that also mean that theres no reason for winbond to be producing bh5 and ch5, since they have a new proccess and it would be a waste if they started producing old chips on an old proccess. :stick:


Technically yes, but they cannot switch to a new process & shut down the old process straightaway.
In anyhow, Winbond stopped production of BH at the end of 2003, so that's an overlap of only abt 6 mths.
What I'm talking abt is it's absurd to run 4 different processes at the same time on similar products, & 2 different processes on identical die design.
It's also implausible for Winbond to invest in both 0.166 & 0.15 when they had 0.13 coming on-stream.

[Revv23]

No need to argue, just read up on their respective datasheets from Winbond.

From the Winbond 942516CH datasheet (for all speed grades from CH-75 to CH-5), the manufacturing process is described as 0.13um. It also says the 942516CH is sorted into 4 speed grades (-75,-7,-6,-5).

I would be very surprised if Winbond has lines that produce ram using 0.18um, 0.166um, 0.15um & 0.13um in an overlapped timeframe. For starters, it means massive investment wasted, as 0.166 & 0.15 are not necessary as Winbond is already migrating to 0.13 at the time when the BH series was in production. The incremental benefits offered by 0.166 & 0.15 over 0.18 does not justify the investment needed. Secondly, why would Winbond waste production cost in producing chips in a larger process when a smaller & more cost effective process is available (not to mention the cost in inventory controls)? Thirdly, what happens to those CH-5 that cannot function at 200MHz but can do so at 166MHz?? Do they junk them altogether?? They surely can't be called CH-6 if it's from a different process!!!

The words AH, BH, CH tells u which generaation it belongs, & the -75, -7, -6, -5 tells u it's speed. If u look at the ICs themselves, it's printed "BH-5" or "CH-5", BH5 or CH5 is just an abbreviation.

http://www.winbond.com.tw/c-winbondhtm/partner/b_2_g_1_a.htm#06

Datasheet for CH series available at bottom of the page.

i wasn't arguing, i was admitting the falseness of my statment. But your statment does indeed put things into perspective for me. Thank you sir.

[clarence]

i wasn't arguing, i was admitting the falseness of my statment.

Sorry, I misread ur reply.

[Rabbi_NZ]

It does make sense for them to produce BH5 right now... the memory market is really hot and I'm sure the likes of Twinmos/OCZ/Mushkin will place large orders for them.
Also, I'm with clarence on the fact that 2 process's at once is nothing new... look at AMD... they are producing both 130nm & 90nm dies RIGHT NOW too.

[situman]

It does make sense for them to produce BH5 right now... the memory market is really hot and I'm sure the likes of Twinmos/OCZ/Mushkin will place large orders for them.
Also, I'm with clarence on the fact that 2 process's at once is nothing new... look at AMD... they are producing both 130nm & 90nm dies RIGHT NOW too.

The difference is that AMD is TRANSITIONING into the newwer process. Whereas Winbond will keep both processes. Correct me if I am wrong.

[annoncompgeek]

Just a comment on dh ch and bh chips.
1)ch-5 was castrated from birth by winbond. It was redesigned during the die shrink to be cheaper to manufacture than bh-5. Cheaper usually means less complex so some of the circutry had to either be redesigned or taken out compleltly. This shows up when comparing bh5 and ch5 at the same speeds and timings. Early ch5 also had an onboard voltage regualtor to limit the vddr seen inside the ram chip to 2.8 . Later chips such as the utt don't have this in their die so scale much better with voltage as proven by the numerous good experiences with it.
2)a smaller die process is more sensative to voltage than a larger one. Even if it can't handle as high of a voltage as a larger process you still get more milage out of it so to speak. If winbond ever makes a dh chip performance will depend on how they design it.
it could be castrated too or it could be the shiznit.
3)infineon actually is liscensing winbond's memory tech and manufacturing capability. Lets not forget infineon leases fab space at amd's hive mind...i mean fab... in germany and is currently refitting for .90 nm production and testing preproduction .65 processes .
4) when amd finally does support ddr2 for a64 i have a feeling ddr1 won't go silently into that good night

[STEvil]

ram chips do not have voltage limitation in them, to do so is suicide due to the amount of heat the excess voltage would generate as it attempted to drain to ground.

[P_1]

so its a seperate ic on the pcb? Could we "destroy" the voltage regulator by feeding imense voltage thru the dimms?(like 3.5 vdimm)

[jjcom]

I wonder...if its on the PCB and we can "see" as in its a little thingy on it, maybe we could remove it. IF someone gets the bright idea that OVP is good.

jjcom

[Revv23]

on CH-5 iirc the v regulation kicked in at 2.8 and cut off past 3.2

[STEvil]

There is no voltage regulation, at all.

[clarence]

The difference is that AMD is TRANSITIONING into the newwer process. Whereas Winbond will keep both processes. Correct me if I am wrong.

They didn't keep both processes. The thing is, they stopped making BH at the end of 2003 (witness someone at Corsair saying they ran out of good BHs sometime during Nov 2003), & I think they announced their substantial exit from the dram market during early 2004.
The recent announcement means they're making new chips with a new process, as evidenced by the need for capital for new manufacturing equipment. Unless Winbond can, in some ways, fire up their BH production line (which is highly unlikely), we are not going to see them making any more BHs.
Winbond won't base their commercial decisions on ppl like us, as companies such as Corsair or OCZ are not their most important customers. In this industry, it's volume that counts.

[NapalmV5]

Is this new BH-5 ?

[enzoR]

yes

[ElrOnD]

when will it be avaiable for sale?

Reviews?:D

[Ref]

They WILL make BH-5 again on the same process ! WOOHOOO :slobber:
:toast:

[enzoR]

no updates alonso?

[metro.cl]

is this official???

[seldomsean]

is this official???

read the thread...

[metro.cl]

read the thread...


thanks for the answer, was it really hard to say yes???

well another question is only twinmos making sticks of these bh5?? and wich pcb are they producing then on???

tccd is still the king of 24/7 use, anything above 3.3v (even at 3.3v) its not nice. to much noise to hot, and on.

so is this new bh5 will oc as high as old ones but give more mhz for the same volts???

[jjcom]

thanks for the answer, was it really hard to say yes???

well another question is only twinmos making sticks of these bh5?? and wich pcb are they producing then on???

tccd is still the king of 24/7 use, anything above 3.3v (even at 3.3v) its not nice. to much noise to hot, and on.

so is this new bh5 will oc as high as old ones but give more mhz for the same volts???

Depends on the person what they say is 24/7 RAM...and nope TwinMOS isn't the only people. OCZ VX uses "CH-5" i'm going to guess that once Winbond gets going again at full we'll be seeing more companies uses their chips.

"Will it overclock higher?" The 512mb sticks seem to be better.

jjcom

[situman]

Although it wouild be nice, Im starting to hear lots of twinmoss deaths due to high volts. Seems kinda fragile for high volt mem. But we will see how the new chips will react once it comes out.

[jumanji969]

Although it wouild be nice, Im starting to hear lots of twinmoss deaths due to high volts. Seems kinda fragile for high volt mem. But we will see how the new chips will react once it comes out.

You expect all memory modules to be perfectly fine at 3.6-3.8v? Thats a lot of extra heat and energy your pumping through them.

[metro.cl]

Although it wouild be nice, Im starting to hear lots of twinmoss deaths due to high volts. Seems kinda fragile for high volt mem. But we will see how the new chips will react once it comes out.


like they say above over 3.5volts is dangerous i you dnot cool de memos the right way and over that you may get lots of deaths remember that people (mosto of them) just put 3.6 or more volts to bench a couple of times and not 24/7 use

[lalPOOO]

I wonder...if its on the PCB and we can "see" as in its a little thingy on it, maybe we could remove it. IF someone gets the bright idea that OVP is good.

jjcom


If there is anything doing something like regulating the voltage, I believe it is possible to slowly destroy it. On my OCZ sticks in sig with the hynix chips on them I've slowly been able to raise the voltage over time. Originally they would give me errors anywhere above 2.9v but now I feed them 3.1v at a lower mhz.

(I'm trying to burn them in a bit, so when my new cooling arrives... I can take them up higher then before)

[BeSaiD]

any news about BH-5?

[Bluebeard]

samples should arrive by next week...

[GazC]

Myth Busted :D
someone should inform Saaya to correct his BH database.

I said this about six months ago and got ignored ;), good luck
:rolleyes:

[enzoR]

np updates :( :stick:

[Ref]

np updates :( :stick:

samples should arrive by next week...

just be patient, enzoR, i think we will have an update soon ;)

[TEDY]

Alonso Gimmmeeeeeeeeeeeeeeeeeee

[enzoR]

just be patient, enzoR, i think we will have an update soon ;)

his "next week" was over 2 weeks ago!

[P_1]

samples should arrive by next week...
err over here it shows 2-03-05... so theres still some time till a week has passed

[ElrOnD]

So is it true that BH5 is back on the market? or just some old sticks?

With the end of TCCD they better be back or we'll have only VX on the market... :eek: :confused:

[Entity_Razer]

hey alonso, quick question, do you have an estimate (wild gues)for how much twinmoss will be selling BH5 equiped bars?

Since they won't be handpicked (except maybe a special SP series or something) I'm assuming the same price as normal SP or twinmos ram but i'd like to know anyways.
:)

[situman]

is it true that too much volts will actually cause instability? I just bought a set of corsair pc3500 with BH5 chips. It seems I cant get it stable at 240fsb with 3.6volts, but im running 245fsb at 3.52volts stable and memtest clean on modded IC7. I thought more volts the better for bh5.

[Revv23]

dp you have the VVT track modded?

if not, then more volts wont help much on IC7

[ZL1]

Wonderful news !!! Thank you OP !!!
Cant wait, cant wait, cant wait :)

Ive been wanting to upgrade my 512mb to 1gb for a while now

by the way guys I hear to run true dual you gotta have 4 sitcks of same size ? is that so ? didnt sound right to me
Id like to keep my 2x256 and add 2x512 (just couldnt bring myself to sell my bh5s)

[Revv23]

4 sticks usuallyt wont run 200mhz, maybe at 2t they might...

[ZL1]

4 sticks usuallyt wont run 200mhz, maybe at 2t they might...

due to ? volts ? I plan to voltmod


Thanks
Dan

[Revv23]

due to a weak memory controller...

[ZL1]

due to a weak memory controller...

:( really wanted to keep the 256s as well

Thanks
Dan

[enzoR]

it will work, you'll get atleast 245mhz 2-2-2 if the sticks can do it, but the chipset will disable PAT.

[situman]

Yea its vtt modded too. Running 255fsb 2228timings. Excellent. Hope the new BH5's will do the same or better.

[BeSaiD]

any news?

[ZL1]

it will work, you'll get atleast 245mhz 2-2-2 if the sticks can do it, but the chipset will disable PAT.

how much would I loose if PAT is disabled ?


Thank you
Dan

[fareastgq]

I don't really see what the big deal is, if it's the same process, it will run just like the ch-5 based sticks. running 1x512 is easier, running 2x512 is hard. I bet the bh-5 average oc will still be about 250-260, lucky ppl will get 275+, crap sticks will be 250-, and at 3.4-3.6 volts is what they will have to be ran at for any type of decent oc. I just don't see them beating out the price/performance ratio of the speed preimium even when they do come out unless they are exactly the same price. (which they won't be if they are binned). but then again, who knows.....nothing like conjecture and speculation, heh.

[saaya]

Myth Busted :D
someone should inform Saaya to correct his BH database.

:doh: so much work and now i gott edit it all over again :D

[Ref]

Alonso, any updates ? :stick:

[P_1]

ok now its been over a week. :confused:

[Skip]

any idea why newegg hasn't been in stock of pc3200 speed premium for like the past month?

[xsky]

if you understand german language you can read about the first new twinmos bh5 here (http://www.forumdeluxx.de/forum/showthread.php?p=1587915#post1587915)

Alex [HWLUXX] seems to have gotten a lot of them. first pair did 250 2-2-2-5 memtest stable with 3.36V

[Bluebeard]

i´ll open a new thread regarding this great news ;)

[zt_lee]

btw, how to differentiate the non UTT twinmos speed premium and UTT twinmos speed premium? any pic to show that?
thks alot

[TEDY]

These 1A4T ram is weird as hell...

240 at 3.0
245 at 3.1
250 at 3.2
and then zip....255 at 3.3 errors...260 at 3.4/3.5v errors

with I should get more don't you think ?

[Rondrian]

it just looks like your cpu is limiting them... the mem controller on it...

[TEDY]

nope proc does 300*9 easily.

And been running my previous BH-5 265-270 also at 3.6-3.7....

[enzoR]

just more burn in ;)

[Jasonhk]

261Mhz @ 3.7v's , no heatspreaders :)
Running on 1st rig in sig

[coop]

Isn't that alot of voltage for 260 - 270, esp. 256 sticks? I mean you can use as much as you want, but if it was not required they may last longer. I'm not saying anything it just seems alot to me.

[TEDY]

I'm disapointed with this TWINMOS SPEED PREMIUM 1A4T....

Best 1 stick does is 250 at 3.2v...the rest is crap.

[enzoR]

hmm i guess we gotta wait for yields to improve. thats probably why twinmos is the first and not many following right away.

[J0lle]

Anyone gotten their Speed Premium PC3500 BH-5 yet?
I ordered 2 sticks for myself yesterday, but they seem to be out of stock, or they never was in stock in the first place.
Im just curious if anyone has gotten to try out these babies? :cord:

[TEDY]

I'm disapointed with this TWINMOS SPEED PREMIUM 1A4T....

Best 1 stick does is 250 at 3.2v...the rest is crap.

2nd and 3rd stick much better :banana:

[Bluebeard]

never leave out of sight what they are rated for!

[TEDY]

tssss alonso if i wanted just 200 at 2.5-3-3-x i would bought some cheap elixir:=)
LOL

[Terry]

let me see if ive got this right. some BH-5 are really good, but some dont perform very well over their rated spec? or they're all good if you give them enough volts?

Anyone seen these for europe -> http://www.oc-wear.com/pd1111585212.htm?categoryId=27

doubt if ill bother as im waiting until that elusive RAM comes along that magically does 270 2-2-2 as then i can up my cpu bus to 270 and lower by multiplier but still max out my cpu :banana: ... in hope 'n' waiting

[enzoR]

let me see if ive got this right. some BH-5 are really good, but some dont perform very well over their rated spec? or they're all good if you give them enough volts?

Anyone seen these for europe -> http://www.oc-wear.com/pd1111585212.htm?categoryId=27

doubt if ill bother as im waiting until that elusive RAM comes along that magically does 270 2-2-2 as then i can up my cpu bus to 270 and lower by multiplier but still max out my cpu :banana: ... in hope 'n' waiting

Charlie's mushkin UTT does 270.

[Revv23]

250 at 3.2v is not good for you?

[Terry]

Cant seem to find the mushkin blue in europe anywhere yet.

i havent gone the 250 route as it's only 18 over what i have now but with a tRas of 2 instead of 3 - i dont think it's enough of a gain to warrant a 120 quid (that's about $220 in conversion), or the vx being around $380 in conversion (£200) :o

[J0lle]

Pictures of Twinmos Speed Premium PC3500 BH-5

got them today

:D

[TEDY]

JOlle test them !!!

[J0lle]

I will, but i have to burn them in first :D

[enzoR]

yea so push 3.2v and see how high they memtest :D
do that for 1h and then 3.3v and so on...

[J0lle]

yea so push 3.2v and see how high they memtest :D
do that for 1h and then 3.3v and so on...

Have to wait a couple of days until i get the booster ;) but after that im rolling, my friend with the same chips got a little over 240 with 3.2V without a little burn-in. The word on the street from a big manufacturer of high quality overclocking-memory said it could take weeks to burn-in these "new" bh-5's properly.. so we will have to see :stick:

[Ailleur]

Anywhere those can be ordered from in north america? ETA?

[enzoR]

Have to wait a couple of days until i get the booster ;) but after that im rolling, my friend with the same chips got a little over 240 with 3.2V without a little burn-in. The word on the street from a big manufacturer of high quality overclocking-memory said it could take weeks to burn-in these "new" bh-5's properly.. so we will have to see :stick:

start memtesting them with the max v's ur mobo can give and run memtest #8. it concentrates on each chip longer than test 5 makeing it better for burnin.

[ianocean]

I've got my new GEIL Ultra X PC-3200 and i was wondering what type of chips is on it because i read is with TCCD but this was the answer of GEIL engineer on GEIL forums:
"GEIL Ultra-X module used to equip with Samsung TCCD. It was able to overclock with low voltage. People love Samsung TCCD.
Unfortunately, Samsung has stopped making TCCD this February. Therefore, we switch IC from Samsung TCCD to Winbond BH-5.
Winbond BH5 needs higher voltage to raise it's clock speed or overclock; standard clock speed require 2.5 to 2.65v, OC to 260 or 270 require atleast 2.85+.
However, there are only few motherboards support 2.85v; most of them are 2.8v top. "

[eva2000]

only way to find out is oc them 2-2-2-5/6 to say 225+ mhz at say 2.9-3.0v vdimm if they pass memtest and windows stability most likely BH-5 dies