5601
Living Room / Re: USB Flash Drives: What to look for or avoid.
« Last post by f0dder on February 27, 2008, 05:58 AM »Do keep in mind that the GT models are plenty faster, that's why you're paying a premium for those.
Recent Posts
, it's cute being able to assign 5 gigabytes to a RAM disk and then manipulate DVD ISOs without going to the harddisk. But heck, under regular use, I don't go much above 2gig that often. Long live overkill!
. Might be able to live with the noise, or I might look for an after-market cooler for it. Also, it seems to draw 20W extra in 2D, I don't dare think how much more it draws when flexxing it's 3D muscles 
- core2 CPUs have a FSB speed of 1066MHz or 1333MHz for the new 45nm CPUs - thus, theoretically, you'd want the RAM to be just as fast. Dual-channeling ram does give a performance boost (although probably not 100% linear scaling), but you might be able to feed a 1333MHz FSB with DDR2-800 in dual-channel mode? But then the large & aggressive core2 caches come into play as well, combined with application memory access patterns, so superfast ram might not even be necessary...) I can't see where a relatively standard workstation user could utilize more than 8GB of memory...Whoo, new GPU! Share the detailsCheap Inno3D GeForce8800GT/512meg DDR3... I considered the new 9600, but it's just about the same price, and performs worse... and not that much lower power consumption.-Lashiec (February 26, 2008, 10:03 AM)
2. I am sending to this customers many games (hugh sized) which are more that 2 GB each.Is it just me, or does this sound just a slight bit fishy?-testgames (December 12, 2007, 09:54 AM)
From my experience using it with video capture from a digital video camera, it isn't. Capturing to a single drive is no problem because the bandwidth required for capture is well below typical drive throughput, (hence where I said a 4200RPM drive is fast enough - if you have problems capturing, the fault generally lies elsewhere, eg. PIO, background processes).Capturing is one thing, but editing, scrolling through frames etc... I would think a stripe makes things a bit more comfortable there? And whether a 4200rpm drive is appropriate for capture probably also depends on the source format... how much throughput does HD video require?-4wd (February 25, 2008, 09:05 PM)
I don't think it would change 'game load speed' but what about loading of data during game play? I'm talking about those games that pause every so often to load in the next >200MB resource file.I thought it would have helped there, since many games seem to have relatively low CPU usage while loading, indicating that they're disk bound. But my RAMdisk try with Thief3 was disappointing, and a "die/reload/try-again/die/..." cycle in half-life2 where everything should be in disk cache (8 gigs of ram...) it was pretty much the same results, so I don't think a stripe would help much there.-4wd (February 25, 2008, 09:05 PM)
When I refer to 'home environment' I don't include a business, (which is what your "whole day's work" implies to me), that runs from home - that's no different from a business in a store or a corporation, (except in size), AFAIC - and as such your backup strategy should be more robust.I know my backup strategy isn't robust enough-4wd (February 25, 2008, 09:05 PM)
When I refer to home environment it is reference to the generic home PC that's used for games, internet, the odd word processing, picture collections, etc. For that, I really don't see any need for RAID. Not even for video editing which I do at home on my general purpose PC.As long as a solid backup strategy is in place, a mirror wouldn't be necessary for that kind of home use. But a backup strategy certainly is, lots of people have irreplacable data on their systems now, and don't even think about the possibility of their drive dying... photo albums, anyone?-4wd (February 25, 2008, 09:05 PM)
Yep, no problem with this for business applications, (and those that are just plain paranoidAgreed. I'm probably borderline paranoid-4wd (February 25, 2008, 09:05 PM)
. But I've experienced losing 3 years of programming and stuff (a mirror wouldn't have helped me then though, but that opened my eyes), and I've been very close to losing a lot of work due to a drive failure.).maybe that is why he has never gotten around to writing a new manual?? Personally, I never understood his mumbo jumbo and actually haven't used the program for a year or two since I could never really verify it did much.Well, to be honest I can't rule out that it does something more than just re-allocating bad sectors. I think his claim is that he "scrubs the sectors with magic bit-patterns to re-magnetize the surface" or something, but I fail to see how that would work - especially since a write to a bad sector triggers reallocation (but OK, perhaps it's possible to turn that feature off temporarily).-techidave (February 25, 2008, 09:39 PM)
Looks like trying the mfg's software may be the best way to go when a drive is going bad. Since we are a small public school in the USA, money is always tight.That's what I would do. And as soon as a drive starts getting re-allocated sectors, I rescue all data and consider the drive as dead; I might use it for transferring non-critical data or as a scratchpad, but not for anything important. I've had drives work like that for years... and I've had them fail a few weeks the first re-allocated sector. Thinking that you have repaired a drive with spinrite (or anything else) is outright dangerous.-techidave (February 25, 2008, 09:39 PM)
During the summer when I usually run MemTest and reghost each student machine, I may have to add the mfg's diagnostic test to this project. Maybe it will save some headaches as the new school year progress.It would probably be a good idea to find a S.M.A.R.T diagnostics tool that can send email reports, and have it trigger on a non-zero reallocated sector count. That'll let you visit potential troublesome machines before things get really bad. If you have routine checks with memtest etc, it's probably not a bad idea adding a disk-vendor-tool full check, just keep in mind that the check itself does put some stress on the system.-techidave (February 25, 2008, 09:39 PM)
I got the file from MS.......
http://www.microsoft...7&displaylang=en
I Googled the file name and it seems to be readily available?-Cuffy (February 25, 2008, 09:07 PM)
This should help boost the adoption of virtual machines. I bet we will see 4 x Quad CPUs with this type of memory. You can pack a lot of computing powers in a little bit of physical space with such a configuration.Yup, Virtual machines could benefit a lot from this, also by having more aggressive disk caching and avoiding harddisk thrashing. And database servers - even moderately large databases could be kept mostly in-RAM (obviously you need some good UPS to prevent loss on power-off-tinjaw (February 26, 2008, 04:37 AM)
).Ok so this brings another question. How does what Spinrite does differ than what WD Diags or anyone's elses do?? Hope you can understand that last question. <grin>This is a very good question, and one I can't answer - simply because Gibson doesn't provide any hard facts, only mumbo-jumbo about magical bit patterns and whatnot. Oh, and spinrite has pretty text-mode graphics including animated sine curves... it does give you an impression it's doing something really profound, doesn't it? Personally I prefer hard facts, rather than tv-shop style circle-jerk praise emails-techidave (February 25, 2008, 08:38 PM)
RAID 1 is actually slower with disk writing than a single disk (though faster on disk read).With good systems, you shouldn't have a really noticable degradation in write speed compared to a single-disk config... and, the flipside, on bad systems you won't see any improvement for read speed. Depends on the RAID implementation. NForce4 sucks, Intel RAID Matrix rocks-Carol Haynes (February 25, 2008, 08:26 PM)
I don't really see the point though as data integrity is the only benefit with RAID 1 and that is easily acheivable in realtime with continuous backup system.Yup, data integrity is the reason you should choose RAID MIRROR, but ~doubling read speed doesn't hurt-Carol Haynes (February 25, 2008, 08:26 PM)
For example, AutoSave intercepts every disk write and mirrors a version of the new/altered file to an alternative location - this gives the same benefits of RAID 1 but adds the benefit of realtime file versioning (you can set it to keep n updated copies of files so you can instantly receive n generations of file changes - nuch more useful than RAID mirroring).Yes, that's certianly cute and you don't get that with a raid mirror. But if the realtime sync/backup is implemented without a filter driver, performance will be hellish for large files...-Carol Haynes (February 25, 2008, 08:26 PM)
. Sure, you'll (currently) be stuck with DDR2-667 speed, but you can get a system with 256 gigs of memory and possibly avoid going to the harddrive for your database..."We had to make our chip look like a DRAM to the memory controller, and like a memory controller to the DRAMs," said Suresh Rajan, the MetaRAM co-founder whom I talked to about the company's technology. This memory traffic routing messes with the DDR2 DRAM timings quite a bit, so the MetaRAM chipset's dynamic command scheduling circuitry ends up doing a kind of "out-of-order execution" with the flow of reads and writes so that the DIMM can operate at a full 667MHz without any glitches.
Whoa! Sounds like I touched a nerve with fodder about Spinrite.Yes, you struck a nerve - I have a real big dislike for false prophets, and I don't want anybody to face a complete drive trashing due to spinrite stress...-techidave (February 25, 2008, 08:07 PM)
