Saturday, January 20th 2024

SSD Overclocking? It can be Done, with Serious Performance Gains

The PC master race has yielded many interesting activities for enthusiasts alike, with perhaps the pinnacle of activities being overclocking. Usually, subjects for overclocking include CPUs, GPUs, and RAM, with other components not actually being capable of overclocking. However, the enthusiast force never seems to settle, and today, we have proof of overclocking an off-the-shelf 2.5-inch SATA III NAND Flash SSD thanks to Gabriel Ferraz, a Computer Engineering graduate, and TechPowerUp's SSD database maintainer. He uses the RZX Pro 256 GB SSD in the video, a generic NAND Flash drive. The RZX Pro uses the Silicon Motion SM2259XT2 single-core, 32-bit ARC CPU running up to 550 MHz. It has two channels at 400 MHz, each with eight chip enable interconnects, allowing up to 16 NAND Flash dies to operate. The SSD doesn't feature a DRAM cache or support a host memory buffer. It has only one NAND Flash memory chip from Kioxia, uses BiCS FLASH 4 architecture, has 96 layers, and has 256 GB capacity.

While this NAND Flash die is rated for up to 400 MHz or 800 MT/s, it only ran at less than half the speed at 193.75 MHz or 387.5 MT/s at default settings. Gabriel acquired a SATA III to USB 3.0 adapter with a JMS578 bridge chip to perform the overclock. This adapter allows hot swapping of SSDs without the need to turn off the PC. He shorted two terminals in the drive's PCB to get the SSD to operate without its default safe mode. Mass Production Tools (MPTools), which OEMs use to flash SSDs, were used to change the firmware settings. Each NAND Flash architecture has its own special version of MPTools. The software directly shows control of the Flash clock, CPU clock, and output driving. However, additional tweaks like Flash IO driving with subdivisions need modifications. Control and Flash On-Die Termination (ODT) and Schmitt window trigger (referring to the Schmitt trigger comparator circuit) also needed a few modifications to make it work.
The overclock results were an increase of 17.6% for the Silicon Motion controller, up to 500 MHz from the original 400 MHz, and 106% with 400 MHz clock for the NAND Flash, which only operated at 193 MHz stock. The performance results? While it varies from benchmark to benchmark, the latency has dropped. Only a slight improvement has been made in reading speeds, while writing has seen a more significant impact. Bandwidth has been increased, so the benchmarks naturally responded well. The biggest problem is the increased operating temperature. Temps were stable at 40C in stock settings, while the overclock pushed it to 45C. Over the long run, this decreases the lifetime of the drive. Overclocking the SSD also voids all manufacturer-implied warranties and reduces the expected number of read/write operations.
Below, you can see a few benchmarks by Gabriel Ferraz, while the rest are available in the YouTube video here.
Source: Gabriel Ferraz (YouTube)
Add your own comment

41 Comments on SSD Overclocking? It can be Done, with Serious Performance Gains

#1
Dirt Chip
Must have direct core LN cooling for my new OC pcie5 NVMe pushed to 2GHz.
Posted on Reply
#2
ExcuseMeWtf
Performance isn't a problem for most SSDs. Most are concerned chiefly with $/GB and long term reliability.
Posted on Reply
#3
qlum
Still if you buy a dirt-cheap sata ssd with the idea of having some fun and overclocking, it does sound like potentially an interesting thing to do.
Posted on Reply
#4
Onasi
It’s a curiosity at best, impressive nonetheless in terms of technical work done. The performance gains would be completely imperceptible in practice, though, and as such - useless.
Then again, this can be said about pretty much every type of overclocking these days. The effort and time involved isn’t worth the result.
Posted on Reply
#5
Quitessa
If you only unlock the full 400MHz of the nand I'd say it's less of an overclock and more of a true performance unlock, or a de-hobbling.
Posted on Reply
#6
hojnikb
Not really news... Ability to change clocks on NAND interface and/or flash controller have existed since forever. These mass production tools have been lying around the interwebs pretty much the moment they became available. More so with flash drives, since there's many more different controllers available too.

I'm surprised it took so long for people to notice this.
Posted on Reply
#7
GabrielLP14
SSD DB Maintainer
Dirt ChipMust have direct core LN cooling for my new OC pcie5 NVMe pushed to 2GHz.
hahaha true
hojnikbNot really news... Ability to change clocks on NAND interface and/or flash controller have existed since forever. These mass production tools have been lying around the interwebs pretty much the moment they became available. More so with flash drives, since there's many more different controllers available too.

I'm surprised it took so long for people to notice this.
True, like i mentioned on the video, intel did a product ages ago, but this went silent over the years, its just as curiosity
QuitessaIf you only unlock the full 400MHz of the nand I'd say it's less of an overclock and more of a true performance unlock, or a de-hobbling.
We could go above it, but the ssd would die much quicker, since its just a curiosity i just wanted to show people
OnasiIt’s a curiosity at best, impressive nonetheless in terms of technical work done. The performance gains would be completely imperceptible in practice, though, and as such - useless.
Then again, this can be said about pretty much every type of overclocking these days. The effort and time involved isn’t worth the result.
exactly
qlumStill if you buy a dirt-cheap sata ssd with the idea of having some fun and overclocking, it does sound like potentially an interesting thing to do.
yeah haha
Posted on Reply
#8
chrcoluk
Storage is one area I wont tinker with, I value my data integrity too much. Those gains I also feel arent game changers either.
Posted on Reply
#9
DeathtoGnomes
I dont see any benefit at all of OC'ing an SSD.
Posted on Reply
#10
lexluthermiester
Since no one seems to mention it, I will: What does this do to NAND durability?

Overclocking other types of electronics is doable because we're generally talking about transistors or other very durable components. NAND flash cells are NOT durable. They are in fact relatively very delicate. So overclocking them? Seems a bit dubious to me. Granted, the article and whatnot above talks about the NAND controller as much as the NAND itself, but still.
Posted on Reply
#11
Gucky
I'd like to keep my SSDs alive for at least 10-20 years, thank you.
They are not getting cheaper... >_>
Posted on Reply
#12
Courier 6
lexluthermiesterSince no one seems to mention it, I will: What does this do to NAND durability?

Overclocking other types of electronics is doable because were generally talking about transistors or other very durable components. NAND flash cells are NOT durable. They are in fact relatively very delicate. So overclocking them? Seems a bit dubious to me. Granted, the article and whatnot above talks about the NAND controller as much as the NAND itself, but still.
yeah, that´s what I was thinking too, I guess that it would have to be tested in normal use to see the effects
Posted on Reply
#13
Onasi
lexluthermiesterSince no one seems to mention it, I will: What does this do to NAND durability?
That’s a good question, Lex, but I assume that in order to actually definitively answer it one would have to OC an SSD and then either use it for as long as it takes for detrimental effects (if there are any) to kick in or just do what AnandTech and Tom’s did way back when they were testing how resilient SSDs actually are (very, turns out) and hit the drive with a scripted full re-write for days until it dies and then compare it to a non-OCd SSD of an exact same model that also underwent the same. In both cases, I think that the guy attempting the overclock wasn’t really thinking THAT long-term. Again, this is all just a curiosity mostly.
Posted on Reply
#14
TechLurker
Brings back memories of shorting pins on Athlons with a pencil to get more performance, and just slapping on a beefier cooler to deal with the heat.

Now to complete the half-meme of OC'ing an SSD, one should slap on the new hybrid AIO SSD cooler from MSI to deal with with increased heat and send it (somehow).
Posted on Reply
#15
trsttte
This is impressive, absolutely useless imo because of the durability implications as others have mentioned but still pretty impressive
Posted on Reply
#16
Count von Schwalbe
lexluthermiesterSince no one seems to mention it, I will: What does this do to NAND durability?

Overclocking other types of electronics is doable because were generally talking about transistors or other very durable components. NAND flash cells are NOT durable. They are in fact relatively very delicate. So overclocking them? Seems a bit dubious to me. Granted, the article and whatnot above talks about the NAND controller as much as the NAND itself, but still.
Gabriel did mention it, not with any data though. This was just over clocked to the rates speed of that particular die anyways.
GabrielLP14We could go above it, but the ssd would die much quicker, since its just a curiosity i just wanted to show people
I am curious to know if you could overclock the controller to support a faster interface, on a PCIe drive. Assuming, of course, the same type of encoding is used.
Posted on Reply
#17
ThrashZone
Hi,
Sure isn't that called mem cache :cool:
Posted on Reply
#18
natr0n
Years ago I mentioned this. Its real now.
Posted on Reply
#19
Tomgang
Cool. But i will not be overclocking a ssd for various reasons. I like to overclocking cpu, gpu, memory and what you can pc. But for ssd with important Files, thats where i set the limits to what i will oc.

First of all. I store important Files on these things. So less reliable is a no go for me.

Secondly with nvme ssd all ready maxing out PCIe gen 4 and if not all ready gen 5, very close to it and with the heat modern nvme ssd can make and the speeds they all ready make. Is oc really needed. For me the answer is no.

My Samsung 980 pro is plenty fast. Even for satan ssd, i can se it makes sense either as a decent Sara ssd all ready maxing out sata 3 speed.

So i will for sure keep my ssd at Stock speeds for reliability, heat and lifespan. I dont need a failing ssd and lost Files. With that said, i have never had a failing ssd since i got my first ssd back in 2011

Off causes for the fun and bragging rights, with an pc ssd. You can say litterly everything is overclock in your pc... Even your storage.
Posted on Reply
#20
mouacyk
I would totally do this, only if the 4K1D rand read increases were worth it. That's where ssd's need the most help.

Caching isn't always reliable, for those that want to suggest it. Even with UPS, you can still have random software/firmware/hardware crashes/reboots and lose the caches. Your os is hosed and needs a rebuild or restore from backup.
Posted on Reply
#21
Jism
He killed the drive, at the end.

Intel was 10 years ago experimenting with such software:


However never released it, due to the complications of SSD overclocking in general.

Dead chips, dataloss, other weird stuff. Imagine the RMA submissions.
Posted on Reply
#22
Scrizz
Count von SchwalbeI am curious to know if you could overclock the controller to support a faster interface, on a PCIe drive. Assuming, of course, the same type of encoding is used.
It depends what you mean by "support." If you mean saturate the bus, that's a no.
It's not just the controller speed that matters but also the speed of the controller - NAND interface and how many NAND channels are available.
Also going to higher clock rates, you have to worry about signal integrity and potentially needing a different pcb substrate.
In addition to those things, you will also need power circuitry that can handle that extra power load without being stressed/overheating.
You're generally limited to between 5-10Watts of power from the host. There's a bunch of other things too, so let's go with no without significant re-engineering.
Posted on Reply
#23
Jism
I am curious tho about those tools used. I have a old Intel 80GB ssd that i would love to toy around with.

Even 10% increase for such drives would be huge.
Posted on Reply
#24
GabrielLP14
SSD DB Maintainer
Count von SchwalbeGabriel did mention it, not with any data though. This was just over clocked to the rates speed of that particular die anyways.


I am curious to know if you could overclock the controller to support a faster interface, on a PCIe drive. Assuming, of course, the same type of encoding is used.
I could, but i didnt had a compatible drive for the MPTools that i had available :/
Posted on Reply
#25
bonehead123
Some folk just seem to have WAY too much free time on their hands :D

I'll just stick with o/c'ing my cpu's & gpu's, at least that doesn't require a bunch of other parts & softwarez to do :roll:
Posted on Reply
Add your own comment
Nov 21st, 2024 07:27 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts