I want to run 8gb!!! in 4 dimms

[beermug]

I recently decided to upgrade from 4gb to 8gb of ocz reaper ddr2 1066mhz memory. running 4gb was no problem, adding 4more caused system lockups,reduced graphics scores, and blue screens.
Since then i have contacted AMD about the PhenomII 2.5ghz triplecore cpu i have. They told me that they successfully run 4 dimms in their test motherboards at 1066mhz without any setting changes. They told me that the memory controller in the cpu might be bad.
Ocz Technologies swares that their Reaper series pc-8500 memory is of the highest quality, even though it needs to run at 2.1volts to produce 1066mhz [so it needs to be overclocked].
Gigabyte technician said that it is very difficult to run 4 dimms at 1066mhz. That most gamers only run 1066 in 2 dimm slots [hence one stick per channel].He also pointed to the memory controller on the cpu as the culprit, and said if i get "true" 1066mhz memory[the low voltage ones that run at 1.8volts], that 4 dimms "might" work at 1066mhz.
So... after many days of research i am frustrated, i would like to run 4dimms because i have Windows 7 64bit operating system. I dont like the idea of overclocking the cpu or the northbridge. I just wanted some extra memory.
I had the idea to look for 4gb dimms that run at 1066mhz and get a pair of those, but i have not found any on the internet so far.
Does anyone have a suggestion in what direction i should go with this? or should i just give it up, and keep running 4gb?

 

[o1die]

I would sell the 710 and go with a propus 620. Newegg has some ocassionally for $85 shipped as open box and I got one recently for $60 at frys.

 

[bilbat]

Would be useful to post MOBO part number...

 

[beermug]

Sorry about that

Gigabyte motherboard MA790GP-UD4H 790GX chipset with HD3300 on board GPU
AMD Phenom II 2.5ghz triple-core CPU
OCZ Reaper series DDR2 1066mhz memory 4GB [2x2gb] @2.1volts

Im beginning to think that what ive heard is true about the cpu/motherboard limitations of running 1066mhz in all 4 slots.

http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/31116.PDF

pages 64-66 seems to have the information on this issue.

 

[beermug]

Would be useful to post MOBO part number...

Sorry about that

Gigabyte motherboard MA790GP-UD4H 790GX chipset with HD3300 on board GPU
AMD Phenom II 2.5ghz triple-core CPU
OCZ Reaper series DDR2 1066mhz memory 4GB [2x2gb] @2.1volts

Im beginning to think that what ive heard is true about the cpu/motherboard limitations of running 1066mhz in all 4 slots.

http://www.amd.com/us-en/assets/co [...] /31116.PDF

pages 64-66 seems to have the information on this issue.

 

[bilbat]

Table 2.8.1 pretty much says it all! What surprises me, however, is that all the GB AMD board manuals used to cite this (1066 limited to 2 DIMMs), usually in both the memspec in the "Product Specifications" list, and the "Installing the Memory" pages - it seems to be missing from your manual! That was specifically why I asked about the p/n - to check the manual...

 

[bilbat]

Oh - and, BTW - THANK YOU!! for the pointer to that documentation; I've been looking for something like that FOREVER! I P&M about AMD as they've not documented WAS, their AGESA start-up code is NDA-only, and haven't been able to track anything down in the way of actual docs - not only the specific doc you list (which appears to contain a lot of the start-up code!), but the 'containing' page will keep me busy for days!

 

[bilbat]

YeeHaahh! ...and the actual word, from the actual prophet's mouth, on the functions of ganged vs unganged memory modes!! (pp 65-66)

 

[beermug]

Actually the credit goes to a guy named "Jae-So" at forums.3dguru.com He is the one that gave me that link to the AMD documentation.
Yea i like that too. I was trying to figure whats better, the guys at OCZ and Gigabyte told me unganged is the way i should go. Looks like they are right cause im running the 2 sticks [one in each channel]. Thats what the motherboard manual recommends.
So it looks like i can add the other 2gb sticks again, and set them at the default of 800mhz instead @1.8v instead of bumping them to 1066@2.1v, or i can get low-latency 800mhz memory instead.
Thanks again to you bilbat and the other guys that assisted me on this.

 

[bilbat]

The difference in raw RAM speed won't slow you down nearly as much as never hitting the swap file will speed you up!

This is from an old post:
To further complicate matters, people often misunderstand the actual quantitative speed improvements inherent in faster ram... Here's the mistake: 1066 is 33% higher than 800 ([1066-800]/800 = 266/800 = .33), so 1066 RAM must be a third faster than 800, right? Not so! You have to figure in latencies. Most 800 will run at 4-4-4-12, while most 1066 is rated at 5-5-5-15, or, even worse, 5-5-5-18. Here's how to appraise the situation in reality: at 800 MHz, a RAM bus cycle is 1.25 μSec long (1000/800); at 1066 (1000/1066), it is roughly .938 μSec long - so, with an 800 stick at a 4 average latency, a RAM bus transaction takes 1.25 (cycle time) times 4 (latency), or 5μSec, while at 1066 it is .938 (cycle time) times 5 (latency), for a transaction time of (roughly) 4.7μSec - so you see, by going to nominally 33% faster RAM, you actually gain three tenths of a μSec per transaction - .3 (transaction gain) over 5(transaction total) = .06, for a real-world improvement of 6%

And this from another:
"There is a place where high speed, versus low latency, will be an advantage - any operations that require large, sustained, reads from and writes to RAM - like, as I mentioned, video transcoding... I always consider my 'pass/fail' system stress test to be: watch/pause one HDTV stream off a networked ATSC tuner, while recording a second stream off a PCI NTSC tuner, while transcoding and 'de-commercialing' a third stream to an NAS media server... But, for the vast majority of people, for the vast majority of use, this is not the case. What's going on behind the scenes: the task scheduler is scurrying around, busier than a centipede learning to tap-dance, counting 'ticks': ...tick... yo - over there, you gotta finish up, your tick is over, push your environment, that's a good fella; oops - cache snoops says we've got an incoherency - grab me a meg for him from over there; ...tick... you - get me the address of the block being used by {F92BFB9B-59E9-4B65-8AA3-D004C26BA193}, will 'ya; yeah - UAC says he has permission - I dunno - we'll just have to trust him; damnit - everybody listen up, we've got a pending interrupt request, everyone drop what you're doing, and you - over there - query interrupt handler for a vector - this is important!!! ...tick.... This is why (aside from the obvious matter of access architecture) that swap files are optimized in 4k 'chunks'... And the most fascinating (scary) thing about it all, is that, at some synaptic, neural level, we're doin' the same thing! (...though, the older I get, the less dependable my interrupt return mechanism is - I repeatedly find myself at the bottom of the basement steps, wondering "now what did I come down here for?!" )

 

[beermug]

hmm trying to take it all in. Alot to learn but im interested. thank you again

 

[beermug]

Best answer selected by beermug.