Slashdot Mirror
Battle of the SATA 3.0 Controllers
Deathspawner writes "Think that all SATA 3.0 (6Gb/s) controllers are alike? As Techgage explores, that's not the case. While most SATA 3.0 controllers do deliver the performance promised, the most popular offering on the market does not — at least where bandwidth-busting SSDs are concerned. The controller comes from Marvell, and was bundled on all motherboards prior to AMD and Intel launching their own SATA 3.0 solutions. In some cases, Marvell's controller is half as fast as the others, making it no better than a SATA 2.0 controller. For those with motherboards using a Marvell controller, the solutions are few; build a new PC, or invest in a super-expensive add-in card."
Can anyone say, "class action lawsuit"? It might not work, but if it's actually promising the performance of the spec and doesn't deliver that seems actionable to me (a legal lay person).
Ever since they got bought by Disney, Marvell's disk controllers were never the same.
There's no -1 for "I don't get it."
I initially tried the Marvell controller on my Gigabyte X58 board for a new Agility 3. It was barely getting 230MB/s reads, and it was capped. It eventually failed to detect the drive, so I tried the Intel SATA 2 controller instead. Not only did the drive detect, but I now get ~250MB/s reads (faster random too I've read). I should've known that the company notorious for their freezing SSD controllers would do no better with the SATA controllers.
Replace MOBO is not a solution?
Oh Noes $50 dollars!
This happens all the time with computers, but especially with drive controllers it seems. The guy who rushes his half baked solution to market first at the lowest price ends up with millions of copies in nearly every computer in the world. Then a couple of years later when people start really using them, they discover that in fact the chip is full of bugs and slow and corrupts your data. It happened with the CMD 640 back when IDE first came out, the SiI 3112 when SATA first came out, and now it's happening against with SATA2. Most early Firewire controllers were total crap too, and the cheap ones still are.
The worst part is that nearly every peripheral card manufacturer is going to use that same chip because it's the cheapest. So even if you try to get around a buggy chip on your motherboard by buying a PCIe card, you'll just end up with a second copy of that broken chip. It's infuriating and I don't expect the situation to change anytime soon. That is why I always wait when a new storage access standard comes out, it's just a solid bet that the first generation chips will be way more trouble than they're worth.
I read the internet for the articles.
Don't be at the front of the technology curve when buying stuff. Let the other guy take the brunt of it all (thank you other guy for testing these things for the rest of us).
I farted and it smelled a bit like vanilla bean extract.
I learned more from that fart than I did from this submission.
Marvell produces bargain chipsets. They work, and that's about it.
If you want performance you get a dedicated part, or at least the standard Intel chip.
Virtually everything JMicron has ever released should probably be mentioned here as well. Those guys really know how to crank up the quality...
Don't buy anything. What the hell do you need SATA 3.0 for? Your single SSD won't be that fast after you've used it a while. As usual, wait six months or a year and there will be much better hardware out there. Don't waste your money now.
If you always wait for the next big thing, you will wait forever. Sometimes you gotta pull the trigger. Personally if I'm comparing two boards equal in all things except SATA 2 versus SATA 3, unless research indicated it was a bad idea, I'd probably go for the more modern variant.
More Twoson than Cupertino
Everyone cheats. If someone doesn't, he should or else his naivety is going to get exploited by everyone else.
No more lawsuits. Take everything you can for free and run with it. Use technology to get stuff for free. That's what technology is for.
Companies don't have morals, why should we have them when dealing with companies? We shouldn't. Companies only care about profits, we should only care about getting stuff for free.
Get even. Play by the same rules as companies do.
Does the quality on a JMicron chip go to 11?
It was the first controller on the market. It's got two ports each rated for 6Gbit/sec and it's connected via a single PCI-E x1 lane that's theoretical maximum is 5Gbit/sec.
Nobody should have been surprised by this at all. The information was readily available.
Basically, it's suitable for a single device that's sata6, and won't outperform the sata3 controller in some areas.
The device was only meant as a stop gap for bleeding edge users to get the capability.
Only when the expected behavior involves going to some other value...
Marvell can be hit or miss sometimes. I remember the issues I had when I built a system for someone that had a 10/1000 ethernet controller built in. I can actually think of a few reasons why the numbers are lower. First of course is the obvious marevell has been sending out the previous generation of chips to manufactures simply to clear old stock. Most people won't notice a difference anyway. If you're already on a SSD, would you really notice a 150mb/s bump? Prolly not. Then again it could simply be a case of a human screw-up, and sending the wrong controller chips out. Another possibility is resilks, and a mishmash of illegal knockoffs.
I'd take the human screwing up and sending out the wrong chips, unless there's some actual proof that this was an order down on high. Or that some company is getting screwed on resilks and knockoffs.
Om, nomnomnom...
I got a vertex 3 ssd (500MB/s read/write) drive a week or so ago. This was after I discovered that my motherboard (Asus P6X58D-E) had a SATA3 controller and I though I could actually use the performance. After I got the drive I did quite a bit of performance tests and discover that the Marvel controller they used is indeed slow. At 4kb the drive does 75MB/s write and 82MB/s read. It peaks at 218MB/s write and 304MB/s read. I switched over to the SATA2 Intel controller instead and at 4kb I got 152MB/s write and 146MB/s read. While it peaks at 271MB/s write and 283MB/s read. There is some drivers at Station-Drivers that are a lot newer than the latest that Asus has but from reading around it might or might not work to flash the inbuilt controller using them. There is also no clear benchmark of what any potential gains would be. For now I have stayed away from them.
Just because you aren't saturating the connection doesn't mean 3.0 isn't worth it. I have a Crucial M4, it'll read up to 415MB/s SATA 2.0 maxes out at 300MB/s; so unless we are talking about a 25% reduction in speed...
From TFA:
This isn't to say that Marvell isn't at a disadvantage, however. AMD and Intel both have internal buses to take advantage of, so their SATA 3.0 solutions are basically unrestricted. Marvell on the other hand has to make use of a PCIe lane in order to get its bandwidth, which for a 2.0 lane is 500MB/s. After overhead, that number effectively becomes around 400MB/s, which is about where we saw the drive's read speed basically cap at.
That kind of writing makes me question the professionalism of Techgage. My God, what a mess. Is he correct? Should I believe his measurements? I really don't know.
"Here Lies Philip J. Fry, named for his uncle, to carry on his spirit"
> For those with motherboards using a Marvell controller, the solutions are few; build a new PC, or invest in a super-expensive add-in card.
$30 to $60 for a Sata 3 controller card on Amazon depending on number of ports and other factors. I haven't built a PC in awhile (I tend to overbuild and then keep them for a long time) but it seems to me that building a new PC isn't *that* cheap, yet.
It's very useful information about the Marvell controller, and I will be watching for that. But the conclusion appears to be hyperbole.
Oliver's law of assumed responsibility: If you're seen fixing it, you will be blamed for breaking it.
I'd be much more interested in a comparison of 4x and 8x SATA controllers, and knowing what kind of throughput they can push. How many can actually saturate their PCIe connection? And of those, which are the cheapest?
Heck, they're only about GBP10-12 in the UK. They ought to be even cheaper in the US.
I want a list of atrocities done in your name - Recoil
It's not always the fault of the controllers, it can also be the way they're connected to the system.
These onboard controllers are connected to the system using PCI Express x1 - it's literally just like plugging them into a x1 slot only they're directly on the motherboard. The problem is there are two versions of PCI Express - the older PCI Express 1.0 provides 250 MB/s in each direction, while PCI Express 2.0 provides 500 MB/s in each direction.
AMD motherboards only had PCI Express 2.0 lanes but Intel had a mix of 2.0 lanes and 1.0 lanes - the most common was 32 x 2.0 lanes (for 2 x x16 lanes for graphics cards) and about 6 x 1.0 lanes coming from the southbridge. So motherboards manufacturers had to either use 1 lane from southbridge and get only 250 MB/s in each direction or resort to using some multiplexing chips that take 2 or more lanes and create a x4 path for the controller. More recently, motherboards detect if there is a card on the second pci express x16 and if there's nothing there, they "borrow" a few of those unused lanes to improve the performance of the various controllers integrated on the motherboard.
See this Anandtech article, it explains better than I can explain: http://www.anandtech.com/show/2973/6gbps-sata-performance-amd-890gx-vs-intel-x58-p55/2
But the point is even if the pci express 2.0 is used, there's only 500 MB/s in each direction, SATA 6 gbps means that a maximum of 750 MB/s should be reachable - very few motherboards connect the controllers to more than one 1x lane so even if the controller could reach 750 MB/s, you won't get it.
This is nothing new - remember the gigabit network cards on PCI? The whole PCI system on your computer can do 133 MB/s and a gigabit link can do about 110 MB/s - would you sue anyone if you plug 4 pci cards in your system and can't reach a throughput higher than 133 MB/s ?
Marvell's controller is half-fast
It gets worse than mere poor performance. Couple NVidia chipsets with Silicon Image SATA and you get corruption.
I use chipset integrated controllers (Intel/AMD) exclusively for SATA and high quality discrete cards for SAS. Even Intel gets this stuff wrong, however. What I don't do is use third party SATA controllers, integrated or otherwise. They always suck. It's a given.
Wait for the standard you want to be integrated into the chipset or be disappointed. The chipset makers put a lot more R&D and validation into their work. That's just how it is.
If the controller isn't living up to advertised specifications then this looks like an opportunity for some lawyer to get rich. Additionally, maybe Marvell will be driven out of business and we won't get saddled with anymore crap from that company!
Does Marvell do anything right ? I know their network interfaces are pretty dodgy, as were their SATA 2.0 kludges.
They know they're a shit company, which is why they rush things to market. Think of all the asian motherboard and add-on manufacturers that are dying to be the first to stick another starburst buzzword on their shiny boxes. Marvell released a shit product a few months before the good ones came out, so they sold millions of chips.
If the manufacturers had any standard of quality, we wouldn't have bottom feeders like Marvell, VIA, Broadcom and friends. Like all other things made in China, it's a race to the bottom. Why should we expect otherwise, when their time is so cheap compared to ours ? If I lose a month's work due to corruption, I'm out a good $5k. If they lose a month's work... well they lost less than the cost of the board.
-Billco, Fnarg.com
is the new super expensive now?
Wow, I must be super rich!
The Kruger Dunning explains most post on
http://www.newegg.com/Product/Product.aspx?Item=N82E16816124046&nm_mc=OTC-Froogle&cm_mmc=OTC-Froogle-_-Hard+Drive+Controllers+/+RAID+Cards-_-Syba-_-16124046
The Kruger Dunning explains most post on
Just purchased an ASUS P8P67 motherboard for a brand new Core i7 2600k install; my first new PC in like 5 years. I chose the P8P67 because it had a good assortment of SATA 3 and USB 3 ports for expansion. I had a DVD burner and 3 SATA drives to put into it; I like lots of storage.
I hooked up the drives, putting my brand new WD Caviar Black 1TB SATA 3 hard drive in the first SATA 3 port, and started installing Windows 7. It seemed to take a long time. The installation finished and I started installing all the usual utilities, apps, and games that one has to install on a new PC. I noticed that my system kept pausing, however. I would try to install something, and I would get frequent hourglass pauses, and sometimes the system would seem to lock up for up to 20 seconds at a time.
Eventually, I looked into the system log and saw that there were a bunch of errors coming up every time this happened; disk unavailable, and a driver name. The driver was for the Marvell SATA controller.
I moved that drive to one of the Intel SATA 3 ports (the other drives were not on the Marvell ports) and I have had no problems of that nature in the three weeks since then.
So, basically two of my SATA 3 ports, one of the primary reasons I chose this motherboard, are of no use to me.
Oh, funny fact; my older system had an ASUS P5N SLI motherboard. Marvell SATA chipset. My and I had both problems with that controller too (identical systems.)
the SiI 3112 when SATA first came out,
WHY WHY DID YOU BRING THAT UP!!! I thought I wiped that bit of my life from my memory. It was traumatic. I still have sleepless nights watching my computer randomly not recognise my 2TB array, and watching the default Linux kernel driver for that POS kernel panic for no reason.
I remember having actual sleepless nights getting that piece of shit to work too. I remember having to install Linux on another drive, then compiling in the appropriate drivers, then using that linux to install linux to my other drives, then chrooting into that new linux to fix things so it wouldn't panic either. (In retrospect there may have been a better way but I was rather new to it at the time).
You should also add RealTek. Any motherboard subsystem with a RealTek chip inside is known to be utter crap. Which is why you will only find RealTek gigabit NICs on mobos targeted at idiots (premium boards with a RealTek chip anywhere? Only idiots buy that) and the very low-end market. Audio is the same, noise levels on the -40dB range are the rule (a proper 5-year old US$ 50 SoundBlaster live delivers noise -90dB).
If you buy anything Marvell, JMicron or RealTek, you deserve the idiots cap. Which would be why I seer clear even from the Intel SSD 510, if it has Marvell inside, it is crap by definition and it will find a way to fuck my day sooner or later, the Intel logo on the cover be damned.
Generally speaking if you want SATA-III to operate satisfactorily you need to use the AHCI controller built into the cpu chipset bundle. That is, the one that Intel and AMD bundle. That will get you a reliable 32-tag-per-port controller. You definitely do not want to use an external controller or a third-party chipset controller (aka Marvell), at least not if you can help it. You won't have a choice if you want hardware RAID, AMD and Intel's controllers don't do RAID (BIOS-based fakeraid doesn't count).
All chipsets have bugs, even AMD and Intel chipsets. Intel AHCI controllers have problems probing Intel SSDs (go figure) and require a driver workaround to unbrick the port when the problem occurs during probe. AMD chipsets don't mask phy errors during initial training, which creates a lot of superfluous interrupts. Both controllers play fast and loose with the AHCI spec and the AHCI spec itself is pretty badly designed, with tons of issues (though not as badly designed as the immensely idiotic USB HCIs).
Another big problem is that the firmware controller that runs the chipset side of the AHCI is typically responsible for ALL the SATA ports, which means that hotplug on one port can actually interfere with operations on another. It pisses me off, but there's no avoiding it.
The external chipsets are even worse. Marvell is a joke. Silicon Image chipsets are full of HARDWARE bugs (not just firmware bugs) which require a lot of workarounds in driver code (for example, you can't abort a soft-reset sequence reliably on a SIL chipset and you can't access the on-chip shared memory while commands are in progress without corrupting any DMA that happens to be occuring).
The stuff is getting better, slowly. The manufacturers of these chipsets have traditionally not really cared about these sorts of bugs because 99.9% of their users are consumers who don't care. The remaining 0.1% professionals who do care aren't a big enough crowd to make the manufacturers actually fix their firmware.
SATA at least has the AHCI spec, too bad more chip manufacturers don't use it. If you want to talk wireless and ethernet chipsets matters are far, far worse.
-Matt (who wrote and maintains DragonFly's AHCI driver)
Trusting a typical performance benchmark is like trusting now many pounds you can bench as a measure of overall health. Yes, benchmarks do provide some information about the relative performance of products, but it's by no means a complete picture!
When you are evaluating performance of a system, it's important to consider all aspects of system performance, and not get too hung up on any single factor.
Recently, we did a server upgrade to using SSDs in our database servers. The performance difference was dramatic, though in benchmark testing, we saw very little performance difference between 1.5 Gbps SATA interfaces and 6 Gbps SATA III interfaces for hosting production databases. We saw something like a 90% drop in system load! This indicates that the biggest benefit of SSDs was in reduced seek time, not transfer rate, and that for our needs, transfer speeds weren't a bottle neck in any event, but that the sharply reduced seek times provided a huge benefit.
Know what you need, and you be much more likely to get it!
I have no problem with your religion until you decide it's reason to deprive others of the truth.
Do lawsuits get your laptop faster ? what a nonsense ...
Also, The Marvell shit controllers, 912x at least, don't pass TRIM commands. It's been proven many times.