Separate names with a comma.
Discussion in 'SSDs & Data Storage' started by nutxo, Feb 19, 2014.
I hated that shit. Is it all ironed out now?
I haven't experienced any stuttering. Not sure what you're talking about.
We arent partial to posts that dont make sense, care to explain it?
SSD used to suffer from "stuttering" . basically things woudl just stop for a second then catch up.
Was this only your experience or was it widespread?
Is there an article about it?
No, I know what the OP is talking about. At a certain point of overtaxing a consumer SATA SD they fell way below what a HD would do. Really annoying. There are a number of explanations that we went over in the past.
I am just torturing a Samsung 840 for the first time and I can't seem to make it misbehave. For now.
I have just killed an Intel 120 GB X-25 with these tests, though, although that one should be been a bit more robust. It wasn't. Very disappointing.
I hate SATA SSDs. Unreliable garbage once you pounce on them enough. And my workloads do.
I don't think you have anything to worry about with the latest gear.
that was back in 2008 ish when they were still new, all the new SSDs are fine and stutter free, best thing to happen to PCs since 3DFX Voodoo
No. Of course the first ones were worse, but no, there is a reason why there are consumer SATA SSDs and professional ones. The standards are different now.
No problems here.
I avoided the first few generations on purpose, but I haven't had any stuttering problems from my SSDs. (Samsung 830 and Vertex 4)
My old Mushkin Chronos Delux 480GB SDD did this. Drove me absolutely nuts. It would perform flawlessly on benchmarks and load incredibly fast as you'd expect. Except, every now and then it would do a 1 second stutter.
For those that haven't experienced it the only way I could explain it is as if you were trying to get up real fast and run full speed, but you trip up on your own feet for a second, recover and haul ass as if "did that really happen?". I'd suggest only a few very keen would ever notice it.
My new Corsair Neutron 512GB drive hasn't had this issue at all, so depending on what you're looking at I think you'll be straight.
I have a stuttering problem with my Gen2 Intel 80 Gbs, I don't notice anything similar with my newer samsung 840 EVO.
I have a wide spread generation gap of Sandforce based SSDs(sig) with no issues.
When I had a WD black on the other hand...
have had a stuttering ssd about 6 years ago, think it was an adata one. it would pause on a lot of writes. none of the new ones do it afaik.
I follow now, the op didnt say how long ago this was.
My brother spent a LOT on early SSDs after believing the hype and found their write performance sucked so bad he vowed to never use another.
Many moons later I showed him a Vertex 2, a year later he bought a Vertex 3 and has been very happy with it.
Yes, the OP was rather useless in description...
The last time I seen this issue was thanks to older OS's (XP & Vista) or those that were not SSD aware and the OS's were putting the drives to sleep. Other times were thanks to poor-arsed controllers having a bad day.
I know what the OP is talking about, it's chronicaled in detail in Anandtech's SSD Anthology. The JMicron stuff was the worst. Basically there were no SSD controller algorithms back then that could correct the inevitable problem of all pages being written to and not being able to delete anything smaller than blocks. The wear-leveling algorithms are great now, with the modern generation starting around Intel's X25-M followed by SandForce's SF2200 models (OCZ was first), Crucial's M4 (using Marvell), and Samsung's 830.
edit: oh and TRIM with TRIM-enabled OS like Win7 helped a ton.
Nah, I can easily make a X25-M hang. All it take is just the right mix of simultaneous reads and then writes.
Can't demonstrate since mine actually died last week.
You musta have worked hard at that. LOL!
I've had 5 of them and they never hung.
The earlier models from any manfg had many flaws (and I had some of them) but Intel's controller and firmware was a major leap forward.
The OP is wayyyyyy behind the times.
How have you guys never heard of this?
First-generation consumer solid state disks had controllers that would choke if they became overloaded by write-requests. All reads would be halted until they could finish servicing a chunk of writes.
To the user, this manifested as the entire system "stuttering" as disk reads were continually paused over-and-over.
No, SSDs do not suffer from this any more, primarily due to improved controllers, programming, and better algorithms for the data-distribution across the NAND chips.
My OCZ Agility 1 60GB from 2010 has this issue if too much data is written at once, but I haven't seen an SSD in the last few years have anything come close to this.
We made posts about it, it would have benefited you to read them
At the start, it wasnt clear if the op was referring to last year, a few years ago or as it turned out the very early days.
SSDs are still far from perfect and can be pretty slow in some circumstances, but fortunately we're long past the days of old, shitty JMicron SSD controllers. I haven't experienced the stutter on my Intel 320, and the 320 is itself fairly old now. As long as you get a SSD with a good controller in it, you're fine. Don't buy any SSD with no-name controllers. Even most of the cheaper drives are less bad than they were a couple years ago, but they are still crap to be avoided. Don't buy a non-HyperX Kingston SSD, for example. Decent drives don't cost much more these days.
I'm still waiting it out for next-gen native PCI-E SSDs (which I've mostly heard should be out in Q2 of this year), myself. Wouldn't really care about upgrading if not for the fact that I have 2 OSes and some games on a single 120GB SSD, heh. Could use the space and might as well wait the little bit longer for the new technology.
I'm pretty sure this is what the OP was referring to.
See references to stuttering here.
While all drives may exhibit strange performance issues here and there on a case by case basis (can't say I have seen any issues with my x25-m 80GB g2 drive, lightning fast), the JMicron JMF601 and JMF602 were infamous for having just terrible random write performance, PERIOD. This was back in 2008.
Various manufacturers in 2008 used the JMF601 and JMF602, including OCZ, superTalent, Patriot, and G.Skill.
I haven't heard of such obvious issues for SSD's since then. None the less, I usually stick with the well reviewed ones that I hear the least issues about.
Samsung 830's and 840's.
Almost any intel.
There are others but those are the ones that come to my mind. Basically, just do a little research first, and don't choose an SSD that hasn't been out at least for a couple of months.
Personally, I'm keeping an eye out for the older intel x25-m 160GB SSD's on ebay. They're old, but they're pretty cheap. Saw a couple had buynow prices of 65$
first gen ssd had some issues, some older motherboard might have but if your on 3 generations MB your fine.
today the ssd works fine.
mine been in work for 3 years now without issue.
3 firmware flashes tho, no issues.
If you have an SSD and you stutter, i would suggest checking your computer for errors.
No, as many others in this thread have stated, the old generation of SSDs did have stuttering problems with the JMicron controllers, which, basically had no data distribution algorithms.
I know because my old OCZ Agility 1 has had this problem when too much data is written to the disk at once.
i know the older ones stuttered, i have first hand experience, but if you have an ssd 2010 and newer i doubt you have stuttering issues. Thats been long fixed. OP was asking for SSDs today.
And OP was answered already, yet you decided to res this thread over two weeks after it was discussed.
You also didn't state that you knew the older ones stutter.
Maybe you should explain yourself rather than putting down a meaningless blurb about something which as been answered, especially not two weeks after the fact.