SK hynix Platinum P51 Launches in South Korea

[TheLostSwede]

Although it was revealed back in March this year, SK hynix new Platinum P51 PCIe 5.0 has only just now launched in South Korea and the specs have been boosted since March. The official sequential read speed is now rated at 14.7 GB/s vs. 13.5 GB/s in March and likewise, the sequential write speed has increased to 13.4 GB/s from 11.5 GB/s. We're also looking at a random IOPS read speed of up to 2,300k and random write IOPS of up to 2,400k. This compares really favourably compared to Phison E26 based SSDs, especially the random IOPS are very impressive.

However, it's not all good news, as the endurance tops out at 1200 TBW on the 2 TB SKU, which is pretty much what every other modern SSD offers today and somewhat disappointing considering that we're looking at SK hynix 238 layer 3D TLC NAND here. The 2 TB SKU is also the largest size available for now and the Platinum P51 will also come in 1 TB and 500 GB flavours. Another not so great, is the 10 W power consumption, even though SK hynix claims it's industry leading, but we know that this isn't true, as the Silicon Motion SM2508 controller is more power efficient than that. The drive has already been quickly benchmarked by South Korean hardware site Quasar Zone and it delivers in terms of the claimed speeds from the manufacturer, at least according to the CrystalDiskMark numbers. No pricing has tipped up as yet, so we'll have to wait to see how competitive the drive will be in an already crowded market. However, it's worth keeping in mind that SK hynix tends to charge a premium for its SSDs over many of its competitors.

Update Dec 14th: The 2 TB SKU has a retail price of 473,000 Korean Won, which equals about US$330, making it prohibitively expensive at the moment, even compared to most other PCIe 5.0 SSDs on the market. The 1 TB SKU comes in at 279,000 Korean Won, or about US$195. The Korean prices include 10% VAT/sales tax.



View at TechPowerUp Main Site | Source

 

[Chaitanya]

Waiting for review.

 

[Nhonho]

One of the problems with PCIe SSDs is that the SSD controllers get very hot because they are manufactured using old lithography to be sold at slightly lower prices.

Controller manufacturers prefer to make them with high electrical consumption rather than making them in more advanced lithography with lower consumption.

 

[Bwaze]

"The 2 TB SKU is also the largest size available for now and the Platinum P51 will also come in 1 TB and 500 GB flavours."

Why downsizing to max 2 TB? Why are we going backwards? How far back will this go?

640Kb should be enough for anyone?

 

[3DVCash]

Finally! Been waiting all year for this!

 

[kondamin]

Is there Anyone working on a plain desktop/laptop (excluding exotic workstation that's running multiple VM's or is being used as an ingest station for some reason) that really needs even faster storage instead of just a lot more?

 

[AnarchoPrimitiv]

I'm a storage nerd, always have been, and I'm sure some of the people reading/commenting here are as well so I will preemptively apologize because they've probably heard me give this diatribe previously, but I want a storage manufacturer to release an all SLC drive for the consumer market. They have them in enterprise and the testing shows they get extremely close to Optane in low queue depth rand reads and writes, not to mention insane endurance. I just wish someone would release some 250GB and 500GB options and at a decent price (though obviously higher than TLC options).

PCIe 5 0 drives really offer nothing compelling to your average user, just as 4.0 drives really didn't either over 3.0 drives....in fact, I'd be willing to bet that absent a large, sequential file transfer, your average user cannot differentiate between the three, but if you've ever had a chance to use a PC with an optane drive as the OS drive, you can perceive a difference (or at least I've convinced myself of that after getting my hands on a discount Intel 905p a few months back).

As it stands now, I feel like the real upgrades in consumer SSDs are more dependent on NAND manufacturers releasing new generations of NAND with increased MT/s (I think the fastest consumers see is 2400 MT/s but I think there is 3200MT/s out there) as opposed to increasing the PCIe bandwidth, either way....I want all SLC drives.

 

[kondamin]

I'm a storage nerd, always have been, and I'm sure some of the people ....

How about micron revives optane?

 

[Bwaze]

Is there Anyone working on a plain desktop/laptop (excluding exotic workstation that's running multiple VM's or is being used as an ingest station for some reason) that really needs even faster storage instead of just a lot more?

I think this has been a question ever sonce second generation of PCIe drives hit the market - benchmarks have had a really big trouble showing any progress besides sequential speeds. All the program startup times, installation times etc are more or less bottlenecked by other things than drives' continuous speed.

 

[Chaitanya]

How about micron revives optane?

you mean 3D X-Point, Optane was Intel's Trademark while Micron had it under 3D X-Point.

 

[Wirko]

They have them in enterprise and the testing shows they get extremely close to Optane in low queue depth rand reads and writes

Even in poor consumer SSDs, free performance is on the table for anyone to pick. Why are OS, application and game developers never expected to learn a thing or two about using queues?

 

[Asni]

One of the problems with PCIe SSDs is that the SSD controllers get very hot because they are manufactured using old lithography to be sold at slightly lower prices.

Controller manufacturers prefer to make them with high electrical consumption rather than making them in more advanced lithography with lower consumption.

This new disk should be based on a 7nm controller as the Sk Hynix Atomos Prime on the Solidigm PS1010/1030, but with more advanced 238L nand flash which is the main limit on sequential speeds.

Solidigm D7-PS1010 1.9 TB Specs

1.9 TB, TLC, PCIe 5.0 x4, U.2

www.techpowerup.com

This is going to make it more efficient than first gen5 disks, not cooler.

 

[TheLostSwede]

This new disk should be based on a 7nm controller as the Sk Hynix Atomos Prime on the Solidigm PS1010/1030, but with more advanced 238L nand flash which is the main limit on sequential speeds.

Solidigm D7-PS1010 1.9 TB Specs

1.9 TB, TLC, PCIe 5.0 x4, U.2

www.techpowerup.com

This is going to make it more efficient than first gen5 disks, not cooler.

It's apparently called Alistar and has a higher model number and different packaging.

SK Hynix Platinum P51 1 TB Specs

1 TB, TLC, PCIe 5.0 x4, M.2 2280

www.techpowerup.com

 

[Asni]

It's apparently called Alistar and has a higher model number and different packaging.

SK Hynix Platinum P51 1 TB Specs

1 TB, TLC, PCIe 5.0 x4, M.2 2280

www.techpowerup.com

Yep, as i said it won't be the same controller but most probably something similar for the consumer market.
Unlike Phison, in the past SK Hynix did not recycle an old gen controller for newer gen disks. That's what we experienced on the E16 (revised E12) and the E26 (E18+ coprocessor).

 

[fynxer]

TBW rating is probably only for warranty purposes.

It usually goes like Warranty valid for 5 year or up to 1200TBW which ever comes first.

If you are heavy duty user they want to escape warranty responsibility as soon as possible and cap you out at 1200TBW for warranty claims.

The true TBW is probably at least double or more for the 2TB.

 

[Scrizz]

I'm a storage nerd, always have been, and I'm sure some of the people reading/commenting here are as well so I will preemptively apologize because they've probably heard me give this diatribe previously, but I want a storage manufacturer to release an all SLC drive for the consumer market. They have them in enterprise and the testing shows they get extremely close to Optane in low queue depth rand reads and writes, not to mention insane endurance. I just wish someone would release some 250GB and 500GB options and at a decent price (though obviously higher than TLC options).

PCIe 5 0 drives really offer nothing compelling to your average user, just as 4.0 drives really didn't either over 3.0 drives....in fact, I'd be willing to bet that absent a large, sequential file transfer, your average user cannot differentiate between the three, but if you've ever had a chance to use a PC with an optane drive as the OS drive, you can perceive a difference (or at least I've convinced myself of that after getting my hands on a discount Intel 905p a few months back).

As it stands now, I feel like the real upgrades in consumer SSDs are more dependent on NAND manufacturers releasing new generations of NAND with increased MT/s (I think the fastest consumers see is 2400 MT/s but I think there is 3200MT/s out there) as opposed to increasing the PCIe bandwidth, either way....I want all SLC drives.

oh like these guys?

SOLIDIGM D7-P5810 800 GB Solid State Drive - 2.5" Internal - U.2 (PCI Express NVMe 4.0 x4) - SSDPF2SQ800GZ01 - Solid State Drives - CDW.com

Buy a SOLIDIGM D7-P5810 800 GB Solid State Drive - 2.5" Internal - U.2 (PCI Express NVMe 4.0 x4) at CDW.com

www.cdw.com

 

[Makaveli]

Want to see power consumption on this device.

Since this looks fantastic from WD

Western Digital Previews M.2 2280 PCIe 5.0 x4 NVMe Client SSDs: 15GBps at Under 7 Watts

www.anandtech.com

 

[TheLostSwede]

Want to see power consumption on this device.

Since this looks fantastic from WD

Western Digital Previews M.2 2280 PCIe 5.0 x4 NVMe Client SSDs: 15GBps at Under 7 Watts

www.anandtech.com

Thanks, I had forgotten to add the power consumption in the news post, it's not great, but I guess it's now awful at 10 W, which I presume is peak power consumption.
Added pricing as well.

 

[TechBuyingHavoc]

oh like these guys?

SOLIDIGM D7-P5810 800 GB Solid State Drive - 2.5" Internal - U.2 (PCI Express NVMe 4.0 x4) - SSDPF2SQ800GZ01 - Solid State Drives - CDW.com

Buy a SOLIDIGM D7-P5810 800 GB Solid State Drive - 2.5" Internal - U.2 (PCI Express NVMe 4.0 x4) at CDW.com

www.cdw.com

That actually is a interesting product. But why is the price so high? I don't mean why is it more expensive than a TLC drive, SLC is always more expensive. But a SLC drive should be only 3 times more expensive based on how many bits can be stored per cell (only on the NAND component of the price), where as this looks to be 20 times higher (assuming a conservative $0.05 per GB for TLC SSDs and roughly $1 per GB for this SLC drive).

A 800GB SLC drive that was available for $120 ($0.15 per GB or 3 times the TLC pricing) would be very attractive.

 

[TheLostSwede]

That actually is a interesting product. But why is the price so high? I don't mean why is it more expensive than a TLC drive, SLC is always more expensive. But a SLC drive should be only 3 times more expensive based on how many bits can be stored per cell (only on the NAND component of the price), where as this looks to be 20 times higher (assuming a conservative $0.05 per GB for TLC SSDs and roughly $1 per GB for this SLC drive).

A 800GB SLC drive that was available for $120 ($0.15 per GB or 3 times the TLC pricing) would be very attractive.

That's not how it works. The SLC NAND is on a different, more expensive node and you don't get as many GB/TB per wafer either, so that means an exponentially higher cost.

 

[TechBuyingHavoc]

That's not how it works. The SLC NAND is on a different, more expensive node and you don't get as many GB/TB per wafer either, so that means an exponentially higher cost.

Ok, thanks. I knew I was missing something here.

Is there no way to make the SLC on a newer node (unless you mean something different by more expensive)? I would think that SLC NAND is simpler to make than MLC or TLC or QLC NAND as it is an older technology.

 

[TheLostSwede]

Ok, thanks. I knew I was missing something here.

Is there no way to make the SLC on a newer node (unless you mean something different by more expensive)? I would think that SLC NAND is simpler to make than MLC or TLC or QLC NAND as it is an older technology.

The problem with NAND is that just like DRAM, it doesn't improve with node shrinks in the same way that we're seeing in something like a CPU or a GPU.
This is true of some other parts that goes into chips as well, such as analogue bits and some RF components.
Right now, the NAND guys are using some kind of 10-20 nm tech to make the NAND dies, but no-one has gone below 10 nm so far and this is part of the reason that they started stacking the NAND dies into 3D NAND chips, as this allowed for several NAND dies to be stacked on top of each other, which lead to greater capacity per chip, as you most likely know.
You could obviously take 3D TLC and use it as SLC, it's done quite often, least not as cache, but you lose out a lot of space and you won't gain that much in terms of performance over a drive that does SLC caching.

The short answer is, no, there are no new nodes that would help make it cheaper to produce SLC NAND.

 

[Scrizz]

Ok, thanks. I knew I was missing something here.

Is there no way to make the SLC on a newer node (unless you mean something different by more expensive)? I would think that SLC NAND is simpler to make than MLC or TLC or QLC NAND as it is an older technology.

To add to what @TheLostSwede said, we must also add the fact that his is an enterprise drive with enterprise ASIC/FW/PLP and the RnD that goes with that.