bh5 steppping database

Quote:

---

Originally Posted by uwackme

6 sticks of KHX3000, D328DW-45 chips.... 4x mixed 0304 0305, and 2x 0304 0307 datecodes on them... 11,2,2,2,2.0;1,3,3,3,3,3,4;eefd;7/15,7/15,7/15 on DFI Infinity/LPuB boards.

3D stable 2 @ 240Mhz, 2@ 235Mhz, and 2@ 230Mhz (0307 chips)

5 of the 6 will do 250Mhz 3D stable tested individually, the 6th does 240Mhz. All 6 will do 250Mhz 2D stable benchs and prime.

---

did you bought them from zipzoomfly ?

all has A00 on sticker ?

Quote:

---

Originally Posted by Electroid

BH5:

Stick 1: "315we" SingleSided KHX3000
Stick 2: "0251PQ1" d328d(v?)-50 BothSided KHX3000 (remarked)

DC async: 2.0-2-2-5 3.64v 264MHz (Kingston KHX3000 2*256mb@NF2) Memtest 1h stable.

Same sticks 2.0-2-2-5 3.7V 274MHz SP1M stable.

---

async?

and the second stick is a double sided 256mb stick? i dont think its bh5 then... maybe bh6? can you try and see what infos aida32 gives you about that stick? please post them and also the infos it gives you about the bh5 stick.

Quote:

---

Originally Posted by Snowman89

DC:
Adata 256 stick:
338WF
227fsb 2-2-2-5 2.9 @ nf7-s poor ocer :(

---

256mb, right?

two sticks and they both do the same and are the same stepping bh5? have you tried them alone?

Async = not 1:1

Asus A7N8X deluxe: CPU/MEM = 3/4

-------------------------

AIDA:

Kingston K
0840EA94h
week 20 / 2003
256 MB (1 rows, 4 banks)
Unbuffered
DDR SDRAM
PC2700 (166 MHz)
64 bit
SSTL 2.5
(7.8 us), Self-Refresh
2.5 (6.0 ns @ 166 MHz)
2.0 (6.0 ns @ 166 MHz)

Kingston K
00077E02h
week 5 / 2003
256 MB (1 rows, 4 banks)
Unbuffered
DDR SDRAM
PC2700 (166 MHz)
64 bit
SSTL 2.5
(7.8 us), Self-Refresh
2.5 (6.0 ns @ 166 MHz)
2.0 (6.0 ns @ 166 MHz)

-------------------------

DS:
KHX3000/256
9905200-017.A00
1122406-2.5V

SS:
KHX3000/256
9905192-019.A00
1213206-2.5V

Quote:

---

Originally Posted by Electroid

Async = not 1:1

Asus A7N8X deluxe: CPU/MEM = 3/4

-------------------------

AIDA:

Kingston K
0840EA94h
week 20 / 2003
256 MB (1 rows, 4 banks)
Unbuffered
DDR SDRAM
PC2700 (166 MHz)
64 bit
SSTL 2.5
(7.8 us), Self-Refresh
2.5 (6.0 ns @ 166 MHz)
2.0 (6.0 ns @ 166 MHz)

Kingston K
00077E02h
week 5 / 2003
256 MB (1 rows, 4 banks)
Unbuffered
DDR SDRAM
PC2700 (166 MHz)
64 bit
SSTL 2.5
(7.8 us), Self-Refresh
2.5 (6.0 ns @ 166 MHz)
2.0 (6.0 ns @ 166 MHz)

-------------------------

DS:
KHX3000/256
9905200-017.A00
1122406-2.5V

SS:
KHX3000/256
9905192-019.A00
1213206-2.5V

---

*cough* i know what async means! :P
i was just asking what rate you were running :D

and i never heard of a dual sided 256mb bh5 stick... thats very very odd... afaik there are only 32mb chips of bh5 and bh6... so i really dont know what to say...

Quote:

---

Originally Posted by saaya

*cough* i know what async means! :P
i was just asking what rate you were running :D

and i never heard of a dual sided 256mb bh5 stick... thats very very odd... afaik there are only 32mb chips of bh5 and bh6... so i really dont know what to say...

---

There were many double sided khx3500 bh-5 (dunno, but IMO even all of them), corsair's 256MB sticks were also double sided... the chips are on both sides to "spread" the heat better and "reduce noise" :rolleyes:

He's right, some KHX sticks were double sided, mainly the 3500, followed by the 3000 and some minor 3200...


P.S. Greetz pal :D

hey, good to see you again! :D did it all work out? new car or did you fix the old old frenchman :D hehe

:doh: hab da auf der leitung gesessen

i had a double sided stick of bh5 myself, the khx 3500 i had was double sided, when i read double sided i assumed it was with 16mb ics :hitself:

LOL stefan, my infinity died! :lol:

system shut down just like that, was just chatting when it went off. never booted again. nothing, the fans didnt even spin up, nada....

tried a different psu, cpu, memory and videocard... nothing... im kinda glad though :D
i will finally go a64 now it seems :)

90nm a64 here i come! :slobber: getting the msi neo2 i think, 4 dimm slots and two memory controllers on the cpu means 4 dimm ocing w/o any trouble! :banana:

so i finally get my 1gb of bh5 with that 4 256mb sticks in one board :lol:

have some question guys please help me out...

how does bh-5 overclock at different vdimm and timings ?

my khx3000/512 so far works like this:

215 2-2-2-5 2.8v
220 2-2-2-5 2.9v

how much to expect on 3-3.1-3.2-3.3-3.4-3.5v etc ?

can bh-5 run like 2.5-3-3-11 at 280MHz or more ?

bh5 usually doesnt overclock higher with lose timings. all the bh5 i tested only oced 5mhz higher with 2.5-3-3 compared to 2.0-2-2 at MOST! ive heard about different stories, but i think in those cases the chipset was limiting them and couldnt run 2.0-2-2 anymore.

bh5 usually gets 2-6mhz from each .1v you give it more. its scales almost linear :) thats why everybody loves it that much :D

if you got 5mhz from 2.8 to 2.9 then you should keep getting 5mhz for every extra .1v you add until the memory maxes out. some does that at 3v some doesnt even stop up to 5v, yepp thats right, 5v vdimm :D dont ask me how long them chips lived but they lived long enough to get some incredible benchmark results :D

if you want to know more ask cpulloverclock :)

so if all goes well you should get

225 3.0
230 3.1
235 3.2
240 3.3
245 3.4
250 3.5

wich is like the average bh5 ive personally owned but 10-20mhz below the average bh5 according to thise database :)

Quote:

---

Originally Posted by saaya

256mb, right?

two sticks and they both do the same and are the same stepping bh5? have you tried them alone?

---

yepp 2x 256mb both same stepping i only tried 227fsb because thats what my mobo could make :( both are 338wf

3200 or 2700?

hmm you're right
looks like scaling is kinda linear

256Mb Kingston ValueRam PC3200
220 2-2-2-6 2.7v
did 244 2-2-2-6 3.25v
and 249 2.5-3-3-8 3.25v
those results done with 10 full loops of memtest and at least 100 loops of test #5

edit: forgot to add
0251we :D

Quote:

---

Originally Posted by saaya

hey, good to see you again! :D did it all work out? new car or did you fix the old old frenchman :D hehe

:doh: hab da auf der leitung gesessen

i had a double sided stick of bh5 myself, the khx 3500 i had was double sided, when i read double sided i assumed it was with 16mb ics :hitself:

LOL stefan, my infinity died! :lol:

system shut down just like that, was just chatting when it went off. never booted again. nothing, the fans didnt even spin up, nada....

tried a different psu, cpu, memory and videocard... nothing... im kinda glad though :D
i will finally go a64 now it seems :)

90nm a64 here i come! :slobber: getting the msi neo2 i think, 4 dimm slots and two memory controllers on the cpu means 4 dimm ocing w/o any trouble! :banana:

so i finally get my 1gb of bh5 with that 4 256mb sticks in one board :lol:

---

Hey, Sascha... we have to meet on MSN again. Too bad I am not up as long as during the semester break :( It's hard to get someone from oversees this way, or even to talk to you ;)

The Peugeot ain't fixed still :(

About the Infinity... hmm. I know these boards sometimes die of some weird causes... but not liking chatting is something new or me ;) Yeah... but you don't seem to regret losing the board too much - dual channel A64 is what I adore too ;) Say, what about the hdds? (Can talk about that on MSN, again)

Cya, my friend :thumbsup:

i took HS off

Kingston
V55875
0305 P01
D328DW-45

so it's BH-5 ? :P

probably... kingston renamed their chips... theres a certain stepping date. before a certain date its bh5, after that date its ch5. i dont know the week, sorry :(

http://pub.lorenz.bei.t-online.de/DW-45/DW45.htm

here you go :)

lol check the small print on the lower end of that page :D

how do i "BURN IN" memory ????

this KHX3000/512 BH-5 :)

burn in means to run it with a high vdimm and using a benchmark to stress it

tedy, please use aida32 and post the infos it gives you, thank you :)

Quote:

---

Originally Posted by ArcTan

hmm you're right
looks like scaling is kinda linear

256Mb Kingston ValueRam PC3200
220 2-2-2-6 2.7v
did 244 2-2-2-6 3.25v
and 249 2.5-3-3-8 3.25v
those results done with 10 full loops of memtest and at least 100 loops of test #5

edit: forgot to add
0251we :D

---

what board/chipset?

Quote:

---

Originally Posted by saaya

burn in means to run it with a high vdimm and using a benchmark to stress it

tedy, please use aida32 and post the infos it gives you, thank you :)

---

you mean like low FSB and high vdimm ?

like 166FSB 2-2-2-5 and 2.9VDIMM ?

Memory Module Properties:
Module Name Kingston K
Serial Number 021A899Ch
Manufacture Date Week 28 / 2003
Module Size 512 MB (2 rows, 4 banks)
Module Type Unbuffered
Memory Type DDR SDRAM
Memory Speed PC2700 (166 MHz)
Module Width 64 bit
Module Voltage SSTL 2.5
Error Detection Method None
Refresh Rate Reduced (7.8 us), Self-Refresh
Highest CAS Latency 2.5 (6.0 ns @ 166 MHz)
2nd Highest CAS Latency 2.0 (6.0 ns @ 166 MHz)

get the max stable speed you can get with 3v vdimm and then run that speed with 3.3v :hehe:

i got max 2.9v :D

damn NF7-S 2.0

with 2.6v i can run it stable 200 2-2-2-5

http://www.softpedia.com/progDownloa...oad-14897.html

try 8rdavcore :D

all my nf7-s 2.0 boards were overvolting vdimm from 2.9v to 3v, are you sure yours isnt? have a multimeter to check?

appx 2.93-2.94v ...no multimeter

i'm using 8rdavcore of course....

my bh-5 is pc3200

Quote:

---

Originally Posted by saaya

what board/chipset?

---

Abit AN7
still testing though I've hit that BH-5 sticks limits at 249 2.5-3-3-8
single channel testing though and default vdd stock board