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

My opinion is that Intel isn't most reliable SSD, Samsung is. Intel is fastest and reliable enough.

 

hehe. why ever you have that opinion. because of the 5 intels worldwide that got bricket? or by the fact that intel took back it's firmware because they didn't want to brick thousands of customers drives, even while they could have made a lot of money selling their devices, while samsung never openly talks about their ssds or firmwares even when you have them on the phone?

samsungs are a joke. they always where (well, the slc ones where nice, but they got one-upped by mtron back in their days).

the intel ones are reliable and get direct support by their company. this is not like samsung, who tries to leave you in the dark.

 

samsung was the pain in my long time ago

 

Facts are:
1st - some Intel users had troubles with 8 MB drives (G1s only, I think)

2nd - two versions of Intel's FW were bricking drives

3rd - Samsung never had any problems with reliability and new FW

4th - Intel is fastest

5th - Samsung gives almost none support

But I was talking about reliability and Samsung SSDs are more reliable then Intel.

 

1st) some users always have problems (there are, if you google around, about, at max, 5 possibly different occurences of that behaviour)

2nd) so did some samsung ones. but unlike intel, samsung never declared what should go where. intel completely blocked the updates.

3th) yes they did. besides not really releasing much firmwares, and not in a public and easy understandable way

4th) yes

5th) yes

6th) samsungs can die. i had one that died in my hands when i tried to install windows vista on it. it's now deco for being my first 600$ spent on an ssd.


7th) this is one thing you forget about intel / samsung: intel only released so far two ssds. samsung quite some more. and not all of them where success (which is why they like to not release them in public so that people put bad stuff into the brand samsung.

here's the one i had first (i have 3 samsungs by now, including this one)



it's more or less jmicron crap style behaviour. terrible.

 

You may be right with your direct experience, no disagreeing there.

What I see though is that any SSD product does not stand on its own. The company behind it and the decisions that company makes is what makes Intel more reliable to me.

Actions are more revealing of intentions than mere words - Samsung offers neither to a large part of it's users. Major fail.

Cheers!

 

If you guys are bored like me, you can test your reboot time using this script. Here's mine (I took the password off as that adds to my time):



~Ibrahim~

 

Dave I feel the exact same way about compression. If I want that, I'll use WinZip, thank you very much

 

This was my first SSD:



And it worked much better then 4800 rpm HDD that I was using before it.

No matter do we like it or not, but Intel have bigger history when it comes to bricked drives then Samsung.

davepemen, I think that this is the first time that someone on this forum is talking about bricked Samsung.

 

no it isn't. and no, intel does NOT have a big history of bricked drives. you just like to believe in that for your own satisfaction.

 

I didn't EVER say that Intel have big history of bricked drives. I said it's relatively big, compared to Samsung's.

 

and this is wrong as you know..

 

I know it isn't, but it doesn't matter, it's obvious that two of us doesn't agree. Which is good, because I'll Google around to fine some cases of bricked Samsung SSDs.

I didn't have any problems with two Samsung SSDs and I hope it will stay the same when my X25-V arrives (the one that sgilmore62 was benchmarking).

 

they will. as there was only one specific case where the intel ssd could actually get bricked: intel gen1 with the original firmwares.

in all other cases, intel reacted fast enough to make sure possible bricking firmware updates can't get accessed in public. there where, other than that g1 case, no ways to brick an intel ssd other than to update it's firmware. for which they took great care of to not happen.

 

OK, now I see where this all is going, now You're telling us when all data from SSD is gone it actually isn't bricked drive.

Just to make myself clear: Intel had much more cases where data was all gone from SSD then Samsung.

And Yes, people should use backups as some of us do.

I won't argue about it any more!

 

The 8MB is related with compatibility problems. It was much more prevalent on Nforce chipset based computers and Apple systems. I guess Apple systems are more proprietary even though it uses a Intel chipset. Then again, some of the Apple systems also use Nforce chipsets.

I'm pretty certain it can be recovered fairly easily.

Of course, the firmware problem is their fault. Even a big company like Intel needs experience in creating a whole new product segment.

 

So....I had some issues over the weekend that required me to use a Seagate 7200.4 instead of my usual OCZ Vertex. Wow. What a wake-up call. Everything felt so sluggish with the 7200.4. Now, two days later, I'm back with my Vertex and everything feels super fast again.

It's not scientific, but I'm certain my Vertex is way faster than my 7200.4.

 

bricked = the drive is not, and won't ever in any way be usable anymore.

and no, intel had not "much more cases" where that happened. as i said, it's mostly internet exaggeration about individual cases. it happened, and it got public, and from then on everywhere got posted "but intel bricks, not?!" and everyone believed thus, that it happened all the time in all sort of cases.

it happened in the beginnings, and not much.

it never happened to gen2 that drives bricked itself during USAGE. they could brick with bad firmware UPDATES, but that's something different. any firmware update on any device can brick. bios updates, hdd firmware updates, etc. always a dangerous thing.

reliability of a drive == does it happen to fail in some form DURING USAGE.

and no, intels don't fail during usage. so don't samsungs.


but you won't understand this point.

 

Is there a $200-ish SSD that's actually good? I just had an OCZ drive's SATA connector break on me, and it's out of warranty.

 

Question for you guys: I was reading on newegg's customer reviews for the Intel X25-M 160GB drive that there was a batch of drives shipped with pre-release firmware and that they were basically defective. Is there any truth to this? I just bought one off eBay so I'm hoping the person that sold it to me isn't offloading one of those.

 

it's on newegg so it has to be true (it's the page where you find bad customer reviews of products not even available yet ).

but in case of ebay, well, if it's written to be FUNCTIONAL on the auction, and you get it and it isn't, ebay will help you to kick his whatever.. so you shouldn't fear. but prepare for a firmware update to the newest firmware anyways.

 

last one:

Intel SSDs had problems after FW update, DURING USAGE after You set password in BIOS for SSD. I won't say anything about second issue with FW because I'm not sure did it died same moment FW is updated or later.

Maybe to buy another SATA connector, unsolder old one and solder new to PCB.

I would try it, it's cheaper then buying new SSD.

 

The BIOS issue wasn't "during usage".

You set a BIOS password and then the drive failed.

During usage would be - "while the OS is booting or running".

The way I understand the BIOS issue that existed, the drive wouldn't even boot if I am not mistaken.

 

You can keep an eye out for Newegg's open box specials. They had a number of X-25M 80gb for $167 but they are sold out right now.

 

That's a pretty tunnel-vision term for the word "usage".....putting a BIOS password on and the drive fails? Yeah, that's pretty normal usage I would say.

I don't know any statistics or anything about any of this, but let's call a failure for what it is.

Just trying to keep this fair.

~Ibrahim~

 

What You want to say? That it's BIOS fault?

So if Your car explode when waiting on red light, then that doesn't count as failed car because it wasn't going more then 25 kmph

LOL

 

The Intel X25M 160gb still beats the Sandforce in some important IOMeter tests because of it's parallelism and 10 channel controller.

http://www.pcper.com/article.php?aid=856&type=expert&pid=7

 

no. during usage a failure means i could be on stage and it just suddenly dies out of nowhere. messing in the bios, setting a password is normally never something you do while doing any relevant work. you do that mostly once at setup time, and then never again. and a simple warning would have been enough, for samsung, for ocz, for anyone. but intel pulled the actual hw out of shops to make sure no one gets hurt.

vertex drives die DURING usage, means you could have bad luck and it just fails while you do an important presentation, or something.

so far, there was only one example of an intel that possibly failed in such a case, and that was the famous overhyped 8mb bug. and even that one was, afaik, during boot, and not while the os was up and running out of nowhere. but i'm unsure about that.

 

All right, fine, I assume your usage patterns are exactly what everyone else's are. So, Sandforce is a cheating drive because it doesn't carry its high performance with your compressed files, but Intel doesn't have failing drives because you don't put a password on the BIOS. So long as it works for you, it must be perfect, right?

Intel is not without it's faults. Let's call a mistake for what it is.

Impossible to speculate these "what if" statements. "If OCZ/Samsung/etc. had drives fail because of a BIOS password, then they would have just put out an advisory notice." You can't ever prove that.

I mean, I would prefer an Intel drive over basically every SSD out there now. But let's just keep this balanced at the same time. I will say I have heard very little in the news about Intel SSDs having problems compared relative to other SSD makers.

 

hello guys, I am having some performance issues with my Samsung PM800 256GB SSD on my XPS1645. I have been running it for a month now, and Windows experience index is down to 5.9 from 7.4. I also noticed that the transfer rates are also slower than before. I also compared this to my other laptop that I sparingly use also with the same SSD, and the WEI is still 7.4. Is there any reason for the performance decline, and any way to fix it? Thanks.

 

no. but changing stuff in the bios is not something you do while you do important work with your ssd. well, at least not normally. nor is it something you do all day long. you do it once, maybe twice in the lifetime of a device. and there was ONE firmware taht COULD then brick the drive, and it was quickly pulled from the market including the drive to make sure no one gets it.

that is NOT something that makes my intel ssds in any form unreliable. because that is a) not a usage pattern fits me (setting and changing bios passwords all day long), and b) not something that would kill my ssd, even if (as they solved the firmware bug, and fixed the issue, and removed the ssd from the marked as long as the fix wasn't there).


an unreliable drive is a drive that can fail while using it. like, now, while i type on my laptop. if it somehow fails right now, that's unreliable. but that won't happen. not with samsung, not with intel. bith ocz vertexes, it can. so i won't buy one of those ever.

 

Post Crystal Disk Info first just to see what FW You have.

 

Here's a question for you guys: Is ~250MBps the real-world saturation point for SATA 3, just as USB doesn't get all the way up to 480Mbps(or w/e unit)? So in other words, with SATA 3, we won't see any drive get over around 250MBps in every category?

 

it's somewhere between 250MB/s and 300MB/s.. i think closer to 270 - 280MB in some specific tests. but more than that, i doubt it. at least not for the real world, where some communication has to happen between disk and os.

but i don't actually know.. i just think i saw some 280MB/s benchmark numbers somewhere?

but yeah, we won't see any drive getting above 300MB/s in every category, and most will be a bit lower at least.

maybe you find some sata 2 RAM-Disk benchmark somewhere?

 

davepermen, I think you misunderstood Mr. Wonderful's question?


I understand it like so: SATA2=300 MB/s of which we can see around 260 MB/s max/sustained.

SATA3=600 MB/s of which we should be able to see 520 MB/s max/sustained.

So, your numbers are basically showing SATA2 speeds - which we can already hit for a while.

 

he asks where the real limit is. because for stuff like usb 2.0, it's not the numerical limit of 480MBit/s

so is the real limit we can see one day on sata2 250MB/s, 260MB/s, 270MB/s, 280MB/s, 290MB/s or 300MB/s.

theoretically, it could be 300MB/s.

i understood his question perfectly well.

but i don't personally know sata communication overhead.

but i guess sata would deliver 300MB/s if you would request a file big enough to not require futher communication (like a 3GB file delivered in exactly 10 seconds, then). but i'm unsure in what chunks the sata protocoll works, so how much overhead happens during fully known big transfers.

 

And, in case I was not clear before... I will NEVER buy a drive that compresses my data. Sorry Sandforce, not for me.

Who will make my future SLC SSD???

 

Did You sell Your Intel or not? I see it's still in signature.

 

trim is just another way to force people to buy new MB-s, new OS and other stuff...

 

haha. indeed. like every other advancement in technique...

it's always just to buy new stuff.

well. get real, learn your stuff.. at least ONE TIME. thanks.


edit: oh, and, there's no need to buy a new MB, or other stuff..
just to clarify that. not even a new os is needed, that's just for automatic trim.

 

Do you guys think I should buy an X25-m 80GB G2 right now, or wait another couple months until those Sata6 SSD's from Micron or whoever, come out?

 

well, i'm one of those who didn't had to wait for some months, and i won't ever regret it. even while it might take a while till i get some sata3 disk + controller.

 

Since I have a notebook without a CD drive, will I be able to update firmware on the Intel G2 from USB stick?

 

should be possible, yes. haven't had the need to update the intels, but others did so with help of a stick afaik.

i updated my mtrons with it. usb-stick-dos was fun

 

New Kingston Drives with TRIM.

On a completely unrelated note, what do you guys think? Should I upgrade from my 80GB G2 to a 160GB G2 now or should I wait for a price drop or refresh? I would think we'd see something when the 320GB comes, which might be soon, seeing as we saw those product listings earlier this month, but who knows.

I would be selling the 80GB to make up the cost of the 160GB.

Another question (which I'm sure I know the answer to): Do you guys see Intel releasing another firmware update ever to improve speeds on the current 160GB drive?

 

if you need the storage, then yes, get it. else, uhm, no. it's only some speed gain in writes. not on the rest of the drive. nothing that would blast your imagination away with amazing new speeds.

and no i guess there won't be another performance related firmware update for them. but i might be wrong. lets wait till sandforce drops

 

Just so you know, I read that Intel will drop prices on the 160 G2 down to around $220 mid-to-late 2010.

 

I like my Kingston V+, but they seem to have jacked up the prices a lot, I wouldn't buy this series for those numbers, it's too expensive.
It will be interesting to see if the firmware is made available for the earlier V+ series and whether the write performance is improved in reality.

 

hello guys, I am having some performance issues with my Samsung PM800 256GB SSD on my XPS1645. I have been running it for a month now, and Windows experience index is down to 5.9 from 7.4. I also noticed that the transfer rates are also slower than before. I also compared this to my other laptop that I sparingly use also with the same SSD, and the WEI is still 7.4. Is there any reason for the performance decline, and any way to fix it? Thanks.

I have attached crystalinfo as requested.

 

That's more than 50% discount on a product that is selling like cup cakes. Where did you read this?

Once the 320GB is released, there will be a decrease in the price of the 160GB, but I dont think it will be that low. If that ever happens, the 80GB should be around $100 or less then