26 Comments(s). 1 Pages(s). Showing page 1. [ 1 ]

   #22. Posted at 09:41 PM on Feb 26th 2008 Edit   Reply

Average mobos can only take upto 4GB max. And believe me installing a 4GB RAM is not that easy. It will result to 3GB and not detecting other module. Some bios setting must be done if supported to be able to detect it. Some folks of THG tested the significant performance of having 8GB in the system (Vista, not XP coz not supported)
Link:
http://www.tomshardware.com/2008/02/15/vista_workshop/

Conclusion:
It will speedup your basic windows appz but not the latest 3D games.
They suggested that 2GB is already a best choice.
collapse

   #1. Posted at 06:51 AM on Feb 26th 2008 Edit   Reply

But are there any benefits to be had for the average PC enthusiast?

Can mobos use/see that much RAM, latencies, etc?
collapse

   #11. Posted at 09:20 AM on Feb 26th 2008 Edit   Reply

Seems like this might be helpful for laptops with only 2 slots.
collapse

   #21. Posted at 04:30 PM on Feb 26th 2008 Edit   Reply

Dude, just use a SIMM tree!
collapse

   #19. Posted at 02:54 PM on Feb 26th 2008 Edit   Reply

This is a flashback. Back in the early days there was a hack for 6800-based machines (early Macs and the Atari STs) that involved physically stacking RAM chips and soldering their pins together by hand.
collapse

   #6. Posted at 08:30 AM on Feb 26th 2008 Edit   Reply

why stop @ $50K for a 256GB memory footprint? drop more cash for a matching 256GB flash drive RAID. :)
collapse

   #2. Posted at 06:53 AM on Feb 26th 2008 Edit   Reply

Holy mother of latency, Batman!
collapse
#2, I laf'd  :   (#9)  «

   #4. Posted at 08:04 AM on Feb 26th 2008 Edit   Reply

Sooo...it's a registered DIMM. Or an alternative to FB-DIMM. Doesn't seem too special except for the power consumption, and the latency penalty that design must cause has to be brutal.

This seems to be nothing but an off-chipset memory controller that happens to be integrated into the module, although with the added feature of responding to the real memory controller with SDRAM signalling.

Wasn't Asus doing something with one mainboard design that used two memory controllers on the mainboard that communicated to the primary chipset's memory controller, in order to give you "quad-channel" memory in some ways? I remember that from maybe a couple of months ago. Or maybe it was Rambus. Seems like that would perhaps be better than this, since at least there's dual-channel throughput from the modules to the memory controller, even if it's still single-channel from that controller to the chipset.

That doesn't look like much "3D chip stacking". It looks like lots of other server memory modules with lots of chips on a module. "Stacking" them vertically along the module isn't "3D".
collapse

   #5. Posted at 08:05 AM on Feb 26th 2008 Edit   Reply

I'm guessing it's not for consumer market, yet. Not for a while.

But damn, this is a pretty big coup for them.
collapse
26 Comments(s). 1 Pages(s). Showing page 1. [ 1 ]
 
Name/Password: / Remember
Reply to:
[click to clear]

[RED] [GREEN]
[BOLD]
[ITALIC] [STRIKE]
[UNDERLINE]

Notice: All posts should abide by the rules, please.
Note: Ctrl-Enter submits the post. (In IE)
DThread keys: Click on a reply to position the blue bar. 'A'/'Z' move it up/down.
Jazztags: (they MUST be closed)
    r{ red }r     g{ green }g     /[ italic ]/     *[ bold ]*
    _[ underline ]_     -[ strike ]-     s[ sample ]s     o[ spoiler ]o  q[ (QUOTE) ]q