• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

PCIe4.0 SSD - Best Options

Joined
Nov 15, 2024
Messages
54 (1.32/day)
In this thread, I was asking specifically about PCIe4.0, 4TB, SSD's, and what people recommend to me. It's written in my OP.

I have another regarding PCIe5.0, as my new build will contain both.

The decision is already made to have both. I'm just trying to identify the actual make and model to get.

Anything other than naming, as per my OP, is pointless to me - though may be of interest to others.
 
Joined
Jun 2, 2014
Messages
561 (0.15/day)
Location
Midwest USA
System Name Core
Processor Intel 12900k @ 5.3/4.0
Motherboard MSI z690 Tomahawk
Cooling Custom H20
Memory GSkill 64GB 3600 cas 15 b die
Video Card(s) ASUS RTX 4070 Super OC'
Storage Optane 900p x2, SK Hynix p41 Pro
Display(s) ACER 250hz 1080p 25" VA display x2
Case Phanteks p500a with all Arctic/Thermaltake fans
Audio Device(s) Focusrite interface, Presonus Studio Monitors and Subwoofer
Power Supply Seasonic 850w plat with cable mod cables
Mouse Glorious Model O
Keyboard Havic mech
Software Win 11 Pro
I prefer SK Hynix p41 pros, they are super fast, the fastest in many categories and are pretty affordable right now. I mostly use Intel Optane 900/905p drives however and they are still faster in real world apps, just not in seq reads/writes.
 
Joined
Oct 5, 2024
Messages
128 (1.58/day)
Location
United States of America
My 2 cents on this is to pick a SSD that has its own software to manage the SSD. I am aware of Samsung and Crucial having such software from personal experience, not sure about anyone else like Kingston. It makes a lot easier to see the statistics of the drive, secure erase or update the drive firmware, etc, etc.
 
Joined
Jul 24, 2024
Messages
301 (1.94/day)
System Name AM4_TimeKiller
Processor AMD Ryzen 5 5600X @ all-core 4.7 GHz
Motherboard ASUS ROG Strix B550-E Gaming
Cooling Arctic Freezer II 420 rev.7 (push-pull)
Memory G.Skill TridentZ RGB, 2x16 GB DDR4, B-Die, 3800 MHz @ CL14-15-14-29-43 1T, 53.2 ns
Video Card(s) ASRock Radeon RX 7800 XT Phantom Gaming
Storage Samsung 990 PRO 1 TB, Kingston KC3000 1 TB, Kingston KC3000 2 TB
Case Corsair 7000D Airflow
Audio Device(s) Creative Sound Blaster X-Fi Titanium
Power Supply Seasonic Prime TX-850
Mouse Logitech wireless mouse
Keyboard Logitech wireless keyboard
Well several youtubers not naming any names has actually tested SATA SSD vs NVME Gen3 and Gen4 and a lot of people thought that the SATA SSD was actually the fastest NVME Gen4 at the time since it felt more fluent when they was using Windows and such.
I, too, have this experience. Windows boots in less than 10 seconds on a old SATA drive, whereas on NVMe drive it can take more than 20 seconds. Tested it on the same computer.
Sometimes I'm really amazed when I see old computer with SATA drive booting Windows much more quickly. Windows' NVMe drivers are not the best.

Right now because Black Friday is coming up I can get the Kingston KC3000, Kingston Fury Renegade or WD Black SN850x in 4TB for the same or less price than a MP700 and is Gen5 speeds that much important?


For most people I doubt.
It only matters with sequential writes/reads.

For DRAM-less drives this mapping table (the index of your files that you are referring to) is still held in memory. It's just your system memory instead of the drive's DRAM.
Now this is where it starts to be interesting. If you plan to OC your RAM, don't you ever buy SSDs without DRAM. Because when your main memory gets unstable or produce errors, there is much more bigger chance that mapping table gets corrupted and thus the whole OS and anything on the drive. Any good DRAM-less drive does have at least small amount of some kind of memory to store the mapping table there instead of system's RAM.

If DRAM-less drives really had a performance cost associated with storing the mapping table in main system memory as opposed to on-board DRAM, it would show in latency test:
It would not, because system memory is much faster and has lower latency than LPDDR4-2444 used in SSDs.

Sustained write speeds are highly dependent on cache size, type of cache used, and how efficient the drive is at using that cache. You can see that MP700 performs about as expected given it's cache size: https://www.techpowerup.com/review/corsair-mp700-elite-2-tb/6.html
DRAM really helps with burst writes.
 
Joined
Jul 13, 2016
Messages
3,338 (1.08/day)
Processor Ryzen 7800X3D
Motherboard ASRock X670E Taichi
Cooling Noctua NH-D15 Chromax
Memory 32GB DDR5 6000 CL30
Video Card(s) MSI RTX 4090 Trio
Storage P5800X 1.6TB 4x 15.36TB Micron 9300 Pro 4x WD Black 8TB M.2
Display(s) Acer Predator XB3 27" 240 Hz
Case Thermaltake Core X9
Audio Device(s) JDS Element IV, DCA Aeon II
Power Supply Seasonic Prime Titanium 850w
Mouse PMM P-305
Keyboard Wooting HE60
VR HMD Valve Index
Software Win 10
Now this is where it starts to be interesting. If you plan to OC your RAM, don't you ever buy SSDs without DRAM. Because when your main memory gets unstable or produce errors, there is much more bigger chance that mapping table gets corrupted and thus the whole OS and anything on the drive. Any good DRAM-less drive does have at least small amount of some kind of memory to store the mapping table there instead of system's RAM.

Stop jumping to conclusions, you honestly think developers of the NVMe protocol didn't think of corruption, unexpected power loss, and transmission errors?

NVMe 1.2 requires HMB (DRAM-less) drives to implement data protection measures to tackle any potential issues:


The mapping table in the RAM is a copy of the one on the SSD. Error correction is used prior to the on-SSD mapping table being altered by data from the HMB in addition to parity bits to prevent errors from data in flight.

DRAM really helps with burst writes.

No, no it does not.

As I've pointed out two times now, the DRAM isn't used as general cache for the entire drive so the misconception that it could provide additional performance is founded on nothing but assumptions. It's used for the mapping table and controller data.

We've also already established that using the main system memory (HMB) vs on-board DRAM isn't the source of notable latency. Given these two factors, where exactly do you expect your theorized performance in burst workloads is coming from?
 
Joined
Dec 14, 2019
Messages
1,210 (0.66/day)
Location
Loose in space
System Name "The black one in the dining room" / "The Latest One"
Processor Intel Xeon E5 2699 V4 22c/44t / i7 14700K @5.8GHz
Motherboard Asus X99 Deluxe / ASRock Z790 Taichi
Cooling Arctic Liquid Freezer II 240 w/4 Silverstone FM121 fans / Arctic LF II 280 w Silverstone FHP141's
Memory 64GB G.Skill Ripjaws V DDR4 2400 (8x8) / 96GB G.Skill Trident Z5 DDR5 6400
Video Card(s) EVGA RTX 1080 Ti FTW3 / Asus Tuff OC 4090 24GB
Storage Samsung 970 Evo Plus, 1TB Samsung 860, 4 Western Digital 2TB / 2TB Solidigm P44 Pro & more.
Display(s) 43" Samsung 8000 series 4K / 65" Hisense U8N 4K
Case Modded Corsair Carbide 500R / Modded Corsair Graphite 780 T
Audio Device(s) Asus Xonar Essence STX/ Asus Xonar Essence STX II
Power Supply Corsair AX1200i / Seasonic Prime GX-1300
Mouse Logitech Performance MX, Microsoft Intellimouse Optical 3.0
Keyboard Logitech K750 Solar, Logitech K800
Software Win 10 Enterprise LTSC 2021 IoT / Win 11 Enterprise IoT LTSC 24H2
Benchmark Scores https://www.passmark.com/baselines/V11/display.php?id=202122048229
Joined
Jul 24, 2024
Messages
301 (1.94/day)
System Name AM4_TimeKiller
Processor AMD Ryzen 5 5600X @ all-core 4.7 GHz
Motherboard ASUS ROG Strix B550-E Gaming
Cooling Arctic Freezer II 420 rev.7 (push-pull)
Memory G.Skill TridentZ RGB, 2x16 GB DDR4, B-Die, 3800 MHz @ CL14-15-14-29-43 1T, 53.2 ns
Video Card(s) ASRock Radeon RX 7800 XT Phantom Gaming
Storage Samsung 990 PRO 1 TB, Kingston KC3000 1 TB, Kingston KC3000 2 TB
Case Corsair 7000D Airflow
Audio Device(s) Creative Sound Blaster X-Fi Titanium
Power Supply Seasonic Prime TX-850
Mouse Logitech wireless mouse
Keyboard Logitech wireless keyboard
No, no it does not.

As I've pointed out two times now, the DRAM isn't used as general cache for the entire drive so the misconception that it could provide additional performance is founded on nothing but assumptions. It's used for the mapping table and controller data.

Article from one of the most popular SSD controller maker:
On an SSD, DRAM is used for several purposes: 1) as a cache when writing data to the drive, 2) as a storage location for indexed metadata and a digital map of where data is being stored across the SSD’s NAND flash memory cells, and 3) as a go-between that handles communication protocols between the SSD and the device’s processor.

...

The main difference between DRAM and no DRAM is performance. DRAM is about 100 times faster than NAND flash memory. That means achieving high performance on SSDs is simpler with DRAM. SSDs without DRAM can still offer high performance, it simply takes a few additional design tricks to achieve it.

The data map stored on NAND in a DRAM-less SSD makes the drive slower (although SSD manufacturers can overcome this issue with intricate SRAM resource utilization and collaboration with NAND), especially when it comes to writing data onto memory cells. That’s because while DRAM can overwrite individual bytes of old data with new information without having to erase it first, NAND flash cannot. With NAND flash, an entire block of data must first be erased before the system can write new data to it.

I don't want to argue with you but when Phison says otherwise as you, I'm with Phison.
When it comes to what DRAM is used for, it really comes down to the firmware. Some disk may rely on DRAM more, some less.

DRAM-less drives might at first glance look as fast as DRAM-equipped, but once (p)SLC cache gets exhausted, sustained and burst writes are way lower than on drives with DRAM.
You can clearly see it here: https://www.techpowerup.com/review/acer-predator-gm7-4-tb/6.html. This, of course, can be mitigated with more (p)SLC cache.

My opinion stays the same - all PC components should be as independent as possible from other PC components.
 
Joined
Nov 15, 2024
Messages
54 (1.32/day)
All - just a gentle reminder.

I'm after:

a) Best-in-class (fastest) - regardless of cost.

and

b) Options that are very-close-to 'Best/Fastest', but, are significantly cheaper.

Specifically, the ID / Name of the item, so that I can search and buy.
 
Joined
Jul 13, 2016
Messages
3,338 (1.08/day)
Processor Ryzen 7800X3D
Motherboard ASRock X670E Taichi
Cooling Noctua NH-D15 Chromax
Memory 32GB DDR5 6000 CL30
Video Card(s) MSI RTX 4090 Trio
Storage P5800X 1.6TB 4x 15.36TB Micron 9300 Pro 4x WD Black 8TB M.2
Display(s) Acer Predator XB3 27" 240 Hz
Case Thermaltake Core X9
Audio Device(s) JDS Element IV, DCA Aeon II
Power Supply Seasonic Prime Titanium 850w
Mouse PMM P-305
Keyboard Wooting HE60
VR HMD Valve Index
Software Win 10
Article from one of the most popular SSD controller maker:


I don't want to argue with you but when Phison says otherwise as you, I'm with Phison.
When it comes to what DRAM is used for, it really comes down to the firmware. Some disk may rely on DRAM more, some less.

Three problems:

1) You ignored parts in the article you linked that you didn't like (or simply didn't read it to completion in a rush to prove your nonsense points, just like your claim that DRAM-less drives should be avoided due to corruption). For example:

Today, however, systems have capabilities such as data (or cache) pre-fetching that enable HMBs to work well and achieve low latency. Again, it’s a bit more of a hassle to get the same performance as an SSD with DRAM, but it’s certainly not impossible.

2) You are misreading the part you are quoting or intentionally leaving parts out:

On an SSD, DRAM is used for several purposes: 1) as a cache when writing data to the drive

Your linked article says nothing about using the DRAM as a cache FOR THE DATA ITSELF. Only that it is used as a cache when writing data and that said cache includes the the mapping table and metadata, exactly as I've pointed out numerous times. It is explained in that very article what the cache is used for:

The DRAM inside the SSD keeps track of where each bit of data is stored as it continually moves around from cell to cell. When data is needed, the SSD controller looks at the mapping table on the DRAM and speedily calls it up and makes it available.

In fact I can link an earlier article from Phison themselves pointing this out:


1732639590322.jpeg


Hey look at that, the cache stores the mapping tables exactly as I've been saying.

In addition, from that same article:

SSDs with HMB enabled generally see significantly less to possibly no degradation in efficiency handling random workloads

Mind you this article is from 2021 and we have clearly reached the point where it's possible to have no degradation in performance using HMB.

Ironically the article goes on to talk about data integrity and how the controller prevents data error and corruption to further disprove your earlier nonsense theory that HMB based SSDs are dangerous for people with overclocked memory.

And because you clearly need more sources on what exactly is stored in the HMB, here are more links that prove my point:


3) Consider that even high-end drives like the 990 Pro only have 1 GB of DRAM per TB (usually, it can be less) and you have a fraction of that available at any time after metadata and the mapping table take up most of it (which is by design, given those are what's supposed to use the DRAM). The idea that massive amounts of bulk data are being stored on high value DRAM ignore that such a thing would likely carry a far greater performance penalty when the controller cannot make room for changes to the mapping table or needed metadata.

DRAM-less drives might at first glance look as fast as DRAM-equipped, but once (p)SLC cache gets exhausted, sustained and burst writes are way lower than on drives with DRAM.
You can clearly see it here: https://www.techpowerup.com/review/acer-predator-gm7-4-tb/6.html. This, of course, can be mitigated with more (p)SLC cache.

What are you even talking about, did you even read that review?

The linked budget DRAM-less SSD is topping the sustained writes chart:

1732637961203.png



Your logic makes zero sense, even if we assume DRAM is being used as a cache for the data itself (it isn't), how in the world does that help in a scenario where cache isn't allowed to recover? You'd fill up the 200 MB of free DRAM (Samsung 990 Pro 1 TB only has 1 GB of DRAM and most of that is going to be used for the mapping table and metadata) in tiny fraction of a second and then there would be zero benefit for sustain workloads and likely performance drops as the DRAM has to clear data out to make room for new metadata and mapping table changes. This is why the DRAM isn't used as a cache, it would be an extremely poor use of a highly value resrouce. In a sustained write scenario where cache isn't allowed to recover the write speed always reverts to the native speed of the flash once cache runs out. That bench is more a test of the controller and NAND, not of the differnces between DRAM and DRMA-less.

In regards to Burst Writes, I've already demonstrated in my last post that all factors that contribute to burst writes can be as good on DRAM-less drives, if not better, than DRAM using drives.

My opinion stays the same

Of course it will and I'm not remotely surprised.
 
Last edited:
Joined
Jul 24, 2024
Messages
301 (1.94/day)
System Name AM4_TimeKiller
Processor AMD Ryzen 5 5600X @ all-core 4.7 GHz
Motherboard ASUS ROG Strix B550-E Gaming
Cooling Arctic Freezer II 420 rev.7 (push-pull)
Memory G.Skill TridentZ RGB, 2x16 GB DDR4, B-Die, 3800 MHz @ CL14-15-14-29-43 1T, 53.2 ns
Video Card(s) ASRock Radeon RX 7800 XT Phantom Gaming
Storage Samsung 990 PRO 1 TB, Kingston KC3000 1 TB, Kingston KC3000 2 TB
Case Corsair 7000D Airflow
Audio Device(s) Creative Sound Blaster X-Fi Titanium
Power Supply Seasonic Prime TX-850
Mouse Logitech wireless mouse
Keyboard Logitech wireless keyboard
Three problems:

1) You ignored parts in the article you linked that you didn't like (or simply didn't read it to completion in a rush to prove your nonsense points, just like your claim that DRAM-less drives should be avoided due to corruption). For example:

Today, however, systems have capabilities such as data (or cache) pre-fetching that enable HMBs to work well and achieve low latency. Again, it’s a bit more of a hassle to get the same performance as an SSD with DRAM, but it’s certainly not impossible.
If you had read my posts correctly, you would have understood that my comments were regarding how DRAM cache on SSDs helps when data is being written, not read.
Now be so kind please and explain how the hell would pre-fetching ever help with written data. And how.

Your linked article says nothing about using the DRAM as a cache FOR THE DATA ITSELF. Only that it is used as a cache when writing data and that said cache includes the the mapping table and metadata, exactly as I've pointed out numerous times. It is explained in that very article what the cache is used for:
I never said that DRAM is used for caching user data. I only said it helps when data is being written. I never said that mapping table is not stored in DRAM.
I'm gonna extend my previous statement and say that DRAM helps with burst writes and sustained writes once (p)SLC gets exhausted, also random writes benefits are not negligible.
It is possible to see that DRAM-equipped drives have dramatically lower amount of speed fluctuations during sustained write test.

Phison explicitly states DRAM purpose for storing metadata and mapping table as another purpose for DRAM:
On an SSD, DRAM is used for several purposes: 1) as a cache when writing data to the drive, 2) as a storage location for indexed metadata and a digital map of where data is being stored across the SSD’s NAND flash memory cells, and 3) as a go-between that handles communication protocols between the SSD and the device’s processor.
There's cache

EDIT: Where the hell did part of my response go?!
Okay, once more ...

There's caching data when writing stated as purpose 1) excplicitly. I don't know what Phison is using DRAM for in purpose 1), but it is something different to purpose 2), what you are always pointing to.
It's up to SSD maker to use that drive's DRAM for any purpose they deem necessary. Of course, DRAM is used for wear-leveling, mapping table, garbage collection, etc, but that does NOT mean it can't be used for anything else.

What are you even talking about, did you even read that review?
I actually did. That drive is doing really great for DRAM-less drive, but only thanks to huge (p)SLC cache. Look how that writing speed fluctuates.
Also, take a look at DRAM-less drive test when such huge cache was not present and compare it to DRAM-equipped drives at that time.

Of course, today's drives are much more advanced than SATA drives back then. NVMe controller is much better at handling this situation now, no discussion about it.
However, this does not mean that the negative impacts of lacking DRAM are still not present in some way even today.
Writing speeds are much more stable, though they may be solid (high) on average as much as with SSDs with DRAM.


Your logic makes zero sense, even if we assume DRAM is being used as a cache for the data itself (it isn't), how in the world does that help in a scenario where cache isn't allowed to recover? You'd fill up the 200 MB of free DRAM (Samsung 990 Pro 1 TB only has 1 GB of DRAM and most of that is going to be used for the mapping table and metadata) in tiny fraction of a second and then there would be zero benefit for sustain workloads and likely performance drops as the DRAM has to clear data out to make room for new metadata and mapping table changes. This is why the DRAM isn't used as a cache, it would be an extremely poor use of a highly value resrouce. In a sustained write scenario where cache isn't allowed to recover the write speed always reverts to the native speed of the flash once cache runs out. That bench is more a test of the controller and NAND, not of the differnces between DRAM and DRMA-less.
My logic says that any non-volatile volatile memory is used basically for caching purposes. And there's a reason why SSD manufacturers still implement DRAM cache on premium SSD drives.
990 Pro 1TB has 1 GB RAM because there's a so-called ratio: 1 MB of RAM for every 1 GB of NAND storage.
Again, how does controller handle DRAM usage and in how many various ways it utilizes the DRAM is defined by firmware.

Going with your logic, I don't understand why SSD makers still make drives with DRAM. What's the purpose if DRAM-less are on the bar with DRAM-equipped?
DRAM is another piece of silicon on the PCB, thus increases costs. It would be wise but unethical move to silentnly remove DRAM with newer model and sell it at same price.

My opinion stays the same - all PC components should be as independent as possible from other PC components.
Because the less things for OS to manage, the better, especially for P'o'S Win11 so called OS.

I think you kind of mismatched my arguments with what iameatingjam was debating with you before.
Now, I'm over with this discussion in this thread, let's not f*ck it up any more. Dude should finally come to a decision on buying a drive.
 
Last edited:
Joined
Jul 13, 2016
Messages
3,338 (1.08/day)
Processor Ryzen 7800X3D
Motherboard ASRock X670E Taichi
Cooling Noctua NH-D15 Chromax
Memory 32GB DDR5 6000 CL30
Video Card(s) MSI RTX 4090 Trio
Storage P5800X 1.6TB 4x 15.36TB Micron 9300 Pro 4x WD Black 8TB M.2
Display(s) Acer Predator XB3 27" 240 Hz
Case Thermaltake Core X9
Audio Device(s) JDS Element IV, DCA Aeon II
Power Supply Seasonic Prime Titanium 850w
Mouse PMM P-305
Keyboard Wooting HE60
VR HMD Valve Index
Software Win 10
I think you kind of mismatched my arguments with what iameatingjam was debating with you before.

I may have, I was under the impression you were saying that the DRAM cache is used for caching bulk files (which it simply isn't large enough for).

If you had read my posts correctly, you would have understood that my comments were regarding how DRAM cache on SSDs helps when data is being written, not read.
Now be so kind please and explain how the hell would pre-fetching ever help with written data. And how.

The reason what you are quoting has nothing to do with writing data is because it doesn't. That's mentioned in the next quote. From what I can gather from your comment, we already agree that the DRAM isn't used for caching bulk data so a lot of these points are no longer in contention.

I never said that DRAM is used for caching user data. I only said it helps when data is being written. I never said that mapping table is not stored in DRAM.
I'm gonna extend my previous statement and say that DRAM helps with burst writes and sustained writes once (p)SLC gets exhausted, also random writes benefits are not negligible.
It is possible to see that DRAM-equipped drives have dramatically lower amount of speed fluctuations during sustained write test.

The uni paper I linked and the Phison article but both point out that HMB has the potential to increase burst writes when using a large HMB. Most drives don't do that as they prefer to simply keep them at peer level with their DRAM-having counter-parts.

I actually did. That drive is doing really great for DRAM-less drive, but only thanks to huge (p)SLC cache.

The Corsair MP600 Elite disproves the idea that a DRAM-less drive needs a large SLC cache to cover up potential downsides. Chart is below.

Look how that writing speed fluctuates.

Yes, I see Corsair MP700 Elite 2TB fluctuates significantly less than many DRAM equipped drives. At the end of the day you are trying to attribute things to the HMB that are due to controller, NAND, and firmware. If HMB is something that causes writing speed to fluctuate you need to be able to demonstrate it across all HMB drives or in a scenario where all variables are account for.

1732725130712.png

1732724779637.png

Also, take a look at DRAM-less drive test when such huge cache was not present and compare it to DRAM-equipped drives at that time.

I'll raise you one, the DRAM-less Corsair MP600 Elite 2TB has a mere 46GB of SLC cache:

1732725810498.png



And yet it's writes are very stable

1732726051598.png


For modern drives with a good DRAM-less implementation controller, NAND, and firmware are much more important than whether the drive is DRAM-less or not.

The linked BX500 is one of the cheapest, crappiest drives on the market. Low performance, endurance, the whole shebang. It'd be worse than me implying the Samsung 870 QVO is representative of DRAM SSD performance. Neither are representative of the performance of DRAM or DRAM-less drives.

However, this does not mean that the negative impacts of lacking DRAM are still not present in some way even today.
Writing speeds are much more stable, though they may be solid (high) on average as much as with SSDs with DRAM.

Ok so you agree that there isn't much of a difference between the two nowadays. In essence you are saying your argument is over small differences.

"negative impacts" Again I'd ask where. Sure you can find old, garbage tier DRAM-less drives that perform poorly but the same could also be said of DRAM-having drives as well. Every metric you've said DRAM-having drives are better at I've proven that in fact DRAM-less can match.

Going with your logic, I don't understand why SSD makers still make drives with DRAM. What's the purpose if DRAM-less are on the bar with DRAM-equipped?
DRAM is another piece of silicon on the PCB, thus increases costs. It would be wise but unethical move to silentnly remove DRAM with newer model and sell it at same price.

The answer to this is simple,

Not all controllers support or do well running DRAM-less and not every company has figured out how to make a good DRAM-less drive yet. Just like everything else new that comes along, technology takes time to mature.

When have you ever seen technology transition on a dime? It almost never happens.
 

Frick

Fishfaced Nincompoop
Joined
Feb 27, 2006
Messages
19,675 (2.86/day)
Location
w
System Name Black MC in Tokyo
Processor Ryzen 5 7600
Motherboard MSI X670E Gaming Plus Wifi
Cooling Be Quiet! Pure Rock 2
Memory 2 x 16GB Corsair Vengeance @ 6000Mhz
Video Card(s) XFX 6950XT Speedster MERC 319
Storage Kingston KC3000 1TB | WD Black SN750 2TB |WD Blue 1TB x 2 | Toshiba P300 2TB | Seagate Expansion 8TB
Display(s) Samsung U32J590U 4K + BenQ GL2450HT 1080p
Case Fractal Design Define R4
Audio Device(s) Plantronics 5220, Nektar SE61 keyboard
Power Supply Corsair RM850x v3
Mouse Logitech G602
Keyboard Dell SK3205
Software Windows 10 Pro
Benchmark Scores Rimworld 4K ready!
Not all controllers support or do well running DRAM-less and not every company has figured out how to make a good DRAM-less drive yet. Just like everything else new that comes along, technology takes time to mature.

When have you ever seen technology transition on a dime? It almost never happens.

So a safe bet is getting a drive with DRAM.

Despite the OPs efforts it has been an interesting thread.
 
Joined
Mar 7, 2023
Messages
921 (1.40/day)
System Name BarnacleMan
Processor 14700KF
Motherboard Gigabyte B760 Aorus Elite Ax DDR5
Cooling ARCTIC Liquid Freezer II 240 + P12 Max Fans
Memory 32GB Kingston Fury Beast
Video Card(s) Asus Tuf 4090 24GB
Storage 4TB sn850x, 2TB sn850x, 2TB Netac Nv7000 + 2TB p5 plus, 4TB MX500 * 2 = 18TB. Plus dvd burner.
Display(s) Dell 23.5" 1440P IPS panel
Case Lian Li LANCOOL II MESH Performance Mid-Tower
Audio Device(s) Logitech Z623
Power Supply Gigabyte ud850gm pg5
Can't we all just agree that dram is something - but not the only thing - that can improve the performance of an ssd?
 
Joined
Oct 5, 2024
Messages
128 (1.58/day)
Location
United States of America
Can't we all just agree that dram is something - but not the only thing - that can improve the performance of an ssd?
We can agree on that. But those other factors are harder to determine for the layperson. It is just easier and quicker to look at DRAM as a marker of SSD quality (even if it is not in some edge cases), because it is either present or not in the SSD.

I find SSDs to be particularly difficult to analyze vs other tech components. All the marketing is focused on sequential speeds, it is the big number on the box, but sequential speeds don't really matter in daily use and even bad drives have pretty good sequential speed these days. IOPS is another metric that is more relevant to daily use but prolonged I/O will change the IOPS you can sustain on a SSD. The amount of free space also changes the speed of the drive. None of this is predictable for performance the way that clock speed or cores on a CPU is.

So when I see no DRAM on a SSD, that is only more complexity for me to figure out. Does this SSD need DRAM or not? Is longevity or performance affected or is DRAM not needed? Having DRAM only hurts energy consumption so I am heavily biased towards SSDs that have DRAM because that is one less complicating factor that I need to keep track of.

It doesn't help that historically, most non-DRAM SSDs have been low-quality cheapo drives and most DRAM SSDs have been higher-quality, higher-performance drives from reputable and long-lived companies like Micron, Samsung, Intel, etc.
 
Last edited:
Top