Talk:I-RAM

Computing Low‑importance

	This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing articles
Low	This article has been rated as Low-importance on the project's importance scale.
	This article is supported by Computer hardware task force.

"i-RAM Rev 1.3 5.25-inch case"

there are no references for this whatsoever, the 1.3 is still the PCI version, gigabyte says nothing about it anywhere on it's website and still shows v1.3 as the pci version.

there is no 5.25" version anywhere to speak of in September 2007

I think the v2 is VAPORWARE as it was announced in 2006 and still no dice.

—Preceding unsigned comment added by Eckre (talk • contribs) 16:04, 13 September 2007 (UTC)[reply]

I have an i-RAM, and it does not take up two PCI slots. If you have an older computer which doesn't support SATA, you will need some sort of SATA converter which could be installed in a PCI slot; however, the i-RAM itself only takes up one PCI slot.

I'm fairly sure the mention of the i-RAM taking up two slots refers to how the RAM sticks may interfere with the space of the slot below and make it unusable. —Preceding unsigned comment added by Acid8000 (talk • contribs) 12:57, 4 April 2008 (UTC)[reply]

external power

has anyone tried using an old computer to power this and link the sata cable to the new computer? would that be feasible? since it only use pci for power, maybe i can solder an old power supply directly to the pci. —Preceding unsigned comment added by 71.255.76.166 (talk) 18:05, 4 September 2008 (UTC)[reply]

Wikipedia is not really the place for this sort of discussion (talk pages are supposed to be about improving the article - probably get more help if ask on usenet or forums on some other website like silentpcreview). Quick thought though - you would want to consider grounding issues carefully. (Make sure the 2 power supplies have common ground). Zodon (talk) 04:44, 5 September 2008 (UTC)[reply]

Advantages/Disadvantages and References

Under "Disadvantages" it says: "High power consumption compared to hard disk" and indicates references 2, 3, and 4.

Number 4 is simply a note and not even a reference, so maybe it shouldn't even be listed under "References" (?)

Number 2 lists powers for 2.5 inch hard drives - presumably referenced as something to compare the RAM power draw to(?)

Number 3 appears to be the only actual reference to RAM/DIMM power usage.. However.. Not only does the referenced article itself state that "The largest measured increase was only marginally larger than the potential error in the system" - but the article also says that they intentionally stressed the RAM modules while they measured them, so even if their results were not blurred by the potential error in the system, they'd still only apply for people who ran Memtest86 on the specific RAM modules 24 hours a day. That also makes the "reference" number 4 (even more?) questionable... since number 4 says: "Assuming DDR RAM takes on the order of 1W/1GB DIMM..." - "assuming..." pretty much means "I'm NOT going to site a reference..." - right..?

Another bit about reference 3 is that the article seems to indicate that they assume that the CPU was the only load on the +12V2 line, so it wasn't listed... but also implies that the CPU does *not* draw any power from the +12V1, +5V, and +3.3V lines - otherwise the measurements will also include power drawn by the CPU during Memtest86. This may be perfectly valid, but I just don't know, and since they didn't specifically say - figured I'd mention it so that if someone who knows better reads this they can clarify.

In any case it seems to leave the whole "Disadvantage" of "High power consumption compared to hard disk" questionable. UnRheal (talk) 22:41, 10 August 2010 (UTC)[reply]

Merger proposal

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.

Merger. -- Alan Liefting (talk - contribs) 01:01, 2 April 2012 (UTC)[reply]

I propose that IRam be merged into I-RAM. I think that the content in the IRam article can easily be explained in the context of I-RAM, and the I-RAM article is of a reasonable size in which the merging of IRam will not cause any problems as far as article size or undue weight is concerned. JanetteDoe (talk) 01:00, 20 December 2011 (UTC)[reply]

Support Per JanetteDoe. SarahStierch (talk) 07:01, 24 January 2012 (UTC)[reply]
Support The two articles are about the same technology and the same product. Unnecessary duplication of content, minimal content, and significant overlap all apply. The information from the IRam article should be integrated into the I-RAM article and then deleted. Listmeister (talk) 15:10, 28 March 2012 (UTC)[reply]

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

intro and usage and flaws

intro should be rewritten so it includes "PCI", it was "RAM on PCI".

Also it should contain why it was produced: high performance ssds were 2 years away, fastest consumer hd on the market, the Western Digital Raptor X wasn't much better than the competition. faster boot devices were wanted. intel core 2 duo conroe was aone of the biggest upgrades for gamers in years but it came with new ddr2 so there was abundance of ddr1 memory. so that's where gigabyte's idea came into place - to reuse memory that would otherwise be thrown away.

Problems: PCI connector cappped out at 133MB/s. PCI is a shared bus. Some old motherboard designs had onboard sound card, network interface with PCI slots running off of those 133MB/s. But it also had SATA port so it allowed full power on that port all the time unless it hit a bottleneck somewhere else usually between southbridge and northbridge. FPGA (very important part) translates between memory controler and SATA. had a battery pack for around 16h of power loss . acess time was 0.0milisec. 8GB of max capacity - only in theory - no support for ecc memory (but ecc server ddr1 modules were made at that time in 2GB capacity) - which equated to poor capacity, poor power loss resiliency, very expensive unless you had excess memory, interface bottleneck. Intel® Optane™ technology kind of has traces of it (ispired by). 89.201.184.8 (talk) 16:14, 26 April 2020 (UTC)[reply]

Your "why it was produced" observation is excellent info. IF this article (and similar) survive(s) and/or if this ground will ultimately be covered by some sort of article on (what I would call the false start of) pseudo-SSDs, then this context would be very useful to have in a relevant article. I have no idea how to source that in a Wikipedia-acceptable way though. —ReadOnlyAccount (talk) 16:31, 6 December 2023 (UTC)[reply]

More pseudo, less spammy?

I have an idea how this article's and the Hyperdrive (storage) article's (long past) hype issues might be addressed:

Maybe what's actually more justifiable is a general article on pseudo-SSD drives, as a class of initially much-promoted but ultimately arguably failed-in-the-marketplace products that had the same idea of using then-semi-novel onboard LiPo batteries with then-cost-competitive DRAMs to build budget ersatz SSDs. I mean, that's a thing that happened, to some significant extent. People tried it. Gigabyte Technologies were a pretty major company to push this, but they weren't the only ones with the same bright idea. I contend that even cul-de-sacs, false starts, false dawns and failure can be notable, and I think it's all of those. Correct me if I'm wrong. I mean, they really tried to push this for a while, but they were noted for using FPGAs, which certainly back then screamed small production run. I also think the pseudo-SSD was plainly a bad idea, and I wouldn't sleep well with my only copy of not-even-that-important data stored on one of these (much worse recovery odds in the event of failure than spinning rust or real SSDs), but there we are. The catch and downside is that "pseudo-SSD drive" is a purely descriptive term I've just thought of; didn't read that somewhere, didn't even try to google it, don't know if that's a thing other people say, or have said. But sometimes purely descriptive is best, and this could kind of capture the most notable aspects of what's otherwise leaning towards ad copy. But obvs, seeing how much seriously legit articles are being fought, I'm clearly not going to fight for this. I'm just throwing this out there. Make lemonade. —ReadOnlyAccount (talk) 05:28, 6 December 2023 (UTC)[reply]