SSD Thread (Benchmarks, Brands, News, and Advice)

for 4k benchmark, atto vs CDM2.2 vs CDM3, atto in 4k offers much higher numbers in both read and write:

 

Hmm....

How new is your drive? I'm getting values around 23MB/s for 4K writes - the first benchmark was 40 odd MB/s too.

But then again I've written over 650GB to my SSD according to the Intel toolbox - and yes I did apply manual Trim.

 

I just ran ATTO an got this...
(Running other stuff too but nothing disc intensive)

 

host writes shows 467G, i tried manual Trim too.

 

Far as I know, Atto does sequential numbers while the 4K in CrystalDiskMark is random. The 14MB/s vs 19-22MB/s number I said earlier were for CDM.

 

Is there any danger in deleting hyberfil.sys by running elevated command prompt, powercfg.exe -h off? I figure with space at a premium on SSD's why not delete it since I do not use hibernate? Deleting hibernation freed up 3gb on my system.

 

no, no danger. except if your laptop one day doesn't have much battery, and you're not around. as it goes to standby first then (if at all) and later hibernates so it can save all the data you where working on without hazzle.

that would be the only possible danger. else, no.

 

And even then, it should rewrite the file as long as there is space.

 

if you have disabled hibernation, no.

 

Ok, that's true.
But if you disable hibernation it can't try to hibernate from standby once the battery is low

 

powercfg.exe /a says no dice to s1 sleep, does that mean TRIM can't work on my system?

 

which is what i said. it will just shut down, killing possible open apps, and letting you possibly lose data.

 

But you sort of know that your laptop will "die" if you disable hibernation and the battery runs out, no?

 

well, yes. but i like to not have to think about that. one day, you drop it into a corner, just for quick, something comes in between and you let it there for hours. and it wasn't even full at the first moment. so it kills all the data..

i like hibernation, giving me a more relaxed experience..

 

That is true its nice to have a safety net - which makes me wonder... is it on or off on my laptop - I have no idea... I think I should check

 

I have hibernation disabled and I just unplugged my laptop without the battery in and it booted back up after selecting "start windows normally" at the "your computer did not shutdown properly" screen.

 

Yepp, but you loose anything that you had in an open programme and that wasn't saved in any form - i.e. RAM content - plus there is the risk of data corruption.

 

Defragmentation cannot be scheduled on solid state drives(SSD's) so Windows must have a way of detecting them probably from winsat disk?
http://windows.microsoft.com/en-US/windows7/Schedule-Disk-Defragmenter-to-run-regularly
If you have a SSD and have run WEI, you will not see C: available when selecting drives in the disk defragmenter schedule configuration window.
"First, a TRIM-supporting OS (e.g. Windows 7 will support TRIM at some point) queries the hard drive for its rotational speed. If the drive responds by saying 0, the OS knows it’s a SSD and turns off features like defrag. It also enables the use of the TRIM command."
http://www.anandtech.com/storage/showdoc.aspx?i=3531&p=10

 

Dumb question. How do I change the amount of space allocated to System Restore on Vista? I have been saved or at least had an easier time using that brain dead piece of software than I care to admit. I'm a believer. That said, I need much less room than what Vista apparently has set already. I don't remember how to set the space on XP (bad) and old reliable Google hasn't yielded anything of note.

 

start>control panel>system>system protection>configure

 

"Be default, Windows 7 will disable Superfetch, ReadyBoost, as well as boot and application launch prefetching on SSDs with good random read, random write and flush performance. These technologies were all designed to improve performance on traditional HDDs, where random read performance could easily be a major bottleneck."

http://blogs.msdn.com/e7/archive/2009/05/05/support-and-q-a-for-solid-state-drives-and.aspx

 

yeah, by dying we only mean it kills the actual windows session. not the laptop but that, in my case, could be bad enough. as i use mine as a portable device (you know, it's a laptop.. ), there can often be the moment where i have to move on quickly, and just close it and put it into the bag. i could have worked on something important, and it could take a day or two till i get back to my laptop again. and by then, all could be lost as the battery ran out.

a lot of apps have some form of data recovery. but hibernation is a 100% recovery. much better (and i prefer it to standby, as the laptop then is fully off, anyways.. once getting used to it, one can't understand how people "shut down" actually )

 

thanks for the pics and yeah, that was the info i always wanted to spread: win7 manages it all itself. don't touch anything

 

someone reads the blogs finally, i can leave, my job is done....



(in my case, it diddn't disable superfetch interestingly... but i just checked, and it's not that agressive at fetching. it only filled half of my ram with standby data)

 

both superfetch and readyboost are not disabled here automatically by win7.

 

then again, i never care. i was from the first second a superfetch lover. on a 4200rpm drive, it made vista dance around any xp. it took ages to boot, but once done, it was so fast thanks to superfetch

and i know it speeds up some of my systems at boot. around 5 secs faster is my quadcore, f.e.

so no reason to change anything

 

sysmain(superfetch) is turned off on my system


On previous installations with VBM18C1Q it would say readyboost cant be used because the system disk is fast enough that readyboost would not provide additional benefit.

 

win7 leaves readyboost enabled, if i disable readyboost, then i can't use intel turbo cache as tmp directory.

 

Windows 7 leaves it enabled or does installing your Intel turbo cache drivers enable it?

 

Question about intel toolbox:
I partition my intel 160G G2 into 3 partition. When i run the latest intel toolbox, the Intel SSD Management Tools only works on my C drive containing windos7 OS, and drive D and E can not run this function.
When I run mamagement Tools on drive C, it can fill up my drive C, then release all the filled up space, then show me status passed. However, When I run the management tools on drive D and E, it shows me "passed" directly without filling up the dirve to full then release the space.
So, any1 here has a solution on this point? All my partitions are primary partitions property.

 

after installing turbo cache drivers, it's set as this. in mechanical HD case, it will be set automatically both enabled. with SSD, readydrive can't be enabled in win 7. hence we can assume win 7 did this.

 

this thing (64G SLC) is in $300+ range:


http://www.xtremesystems.org/forums/showpost.php?p=4098411&postcount=707
http://www.ssdshop.ch/cat159.htm

 

nice find. cheap for slc.

 

That is a nice find and I'm sure the performance is worth it. However, didn't Tom's Hardware say the power consumption on Solidata SSD's was worse than a desktop Raptor?

 

this ssd has rated power consumption as 3W while raptor got 8-10W:
http://techreport.com/articles.x/9364
tomhardware referred to X1 while this one is K5:
http://www.tomshardware.com/reviews/256gb-samsung-ssd,2265-5.html

 

Waiting for a video (ZDF Mediathek) so I ran a couple of random benchmarks...

Unistalled the Intel Matrix Storage manager, deleted the iaStor key in the registry and installed the again this morning...

No recent trim.

 

well, on a desktkop, i still prefer such a biest over a raptor, even if it eats 10x as much power faster and silent. oh, and, have i said it before? FASTER.

 

I feel stupid but I don't see an option for configure there.

Ah ha. Ok, Myth Busted. Looks like Soliddata managed to up the ante on their controller. While probably more suited for desktop use, it's not an unreasonable choice for laptop use. I think. I remember a while ago a poster commented on the X-25's power consumption and wondered what would have happened to the max performance if Intel didn't care about battery life.

 

Working out actual power consumption of an SSD vs HDD isn't as straight forward as comparing idle and load numbers, you also have to factor in how much time each will spend at idle or load, in the case of an SSD this is likely to be alot less then the HDD as it completes its transactions quicker and returns to idle where a HDD takes longer and so spends more time at peak power use. Compared to the power draw of the CPU & GPU the difference really is small potatoes.

 

Hey i got some question i dont know if im alowed to post it here
what is the brand of Toshiba A505 S6979 SSD?
is it good SSD the toshiba got?
because that time i remember i read a book at library it said that some SSD doesnt support some program/something that is used to directly delete the slot at the SSD not only shown at the computer..

sorry if u didnt understand

 

Do you mean things like Trim, "Garbace Collection"?
The fact that flash modules aren't technically "cleared" when things are deleted?

(without Trim, Garbage collection)

 

yeah i think that one..the one when we erase our data..and actually the SSD not yet really empty the space..but in our computer it shows it deleted and u got increase in free space

 

Well, that's Trim most likely.

And your type of SSD... try CrystalDiskInfo if it will tell you...it might...

 

Yeah, that was me. Like everything in life; balance is what makes things 'great' for most people, not just some.

 

except that in a lot of cases, it makes things just good for all, instead of great for most.

 

anand said this anand said that...
well, anand also said that sammys don't have fw updates while mine sammy was fw updated like a half year before he wrote that stupid lie
so, I don't take his articles to seriously
and, as I said before, ANYTHING that OS or user must take care of on any piece of hardware is not good thing, and, from that point of view, for me, "GC" is much better than TRIM
also, if some kind of official documentation is out for trim, I don't see why it should not be the same for "GC", but, there is not, and until it is out, anand is not the one that will tell me how it works

 

he never talked about SLC sammies. and you know that, you just don't want to state in public that this is true, so you would make your point invalid

well, trim makes the os and the ssd talk 100% with each other. this is obviously a great thing. but indeed, an ssd should be able to manage itself without trim, too. but having trim is, just from logic, essential anyways. they should be able to communicate with each other, no?

get over yourself and accept that, what anand told, still is true. he just never talked about your case.

 

anand said sammsung, not mlc, not slc, he said samsung
at that point, there were no flashers for sammys, that's true, but, he also said that there will never be flashers for samsungs, and that, as time shoved, was just one big lie

 

Please tell me where is a hardware review site that is right about everything 100% of the time. I would like to know.

 

darQ96,

Nobody is telling you what to believe or disbelieve just because they are saying it.

What your stance is on this subject is very obvious though, with no help to others reading this thread or to yourself, I may add.

Don't take anyone's articles seriously, but at least consider the facts presented. Or, present your facts to the opposite - without your obvious biases.

With regards to GC - TRIM is far superior precisely because of the dependance to the O/S. Or, you do want your SSD drive to tell you what is valid data and what is not? At least now, we tell the O/S and the O/S tells a TRIM enabled SSD what we want it to do.

Just because the end results happen to be the same (keeping performance up as high as possible) does not justify the means. You can argue this all you want, but it will not change this simple fact one iota.

Cheers!