Optane 1600X 118GB - Lots of CDM benching and some thoughts

[QcRMoD]

I don't think you can come to conclusions on reality of a different product by reviewing one product.

Your experience of older systems being speedier is simply because their architectures are slower, but more basic, less feature ridden but with lower latency and simpler software, as you've acknowledged. Complex software and wide architectures don't do well at feeling snappy.

Zen 4 needs a minute to train the RAM each boot, this is completely unrelated to whatever the SSD drive is doing, in fact the drive isn't even loaded at that point.

You're right that it's hard to stress the actual performance of Optane in typical consumer usage, but that's kind of the point of the drive, you won't be bottlenecked by it, nor will it wear out within the useful life of the computer it's connected to.

Where this stuff shines is in server applications, hence that's the target market, and you don't see these drives sold to consumers, not really.

I just feel the need to say I entirely agree with this. A SN850X was by no way faster on an accordingly 4.0 platform. If anything, and if not especially with BOOT/REBOOT/WAKE UP, they were all slower. None of my customers or friends would see this. I will see this because I install Windows 6-8x a year either for myself or for them. I AM THE ONLY ONE that will see this, out of all of them.

And my friend that I've rebuilt, or his friend that I've built anew from far lesser specs and HDD reality, have NOTHING to complain about of course. Basically, they never need to reboot Windows 20x like I do when installing and refining their system. They don't neither have to wait through the bulk of a fresh OS install updates. Nor do they have to download and install most of the basic apps they use. I do all of this for them. Nor do they need to get rid of all the shit Win 10 and especially Win 11 will put onto them.

From there, what they DO NEED to do, and how often they DO NEED to reboot... is not something they can even feel or calculate. So then it's mostly waking up, and as much as old systems in DDR3/Sata SSD are a bit faster there, it's not like booting or rebooting. It has no REAL WORLD impact on them. They don't do most of the shit I do EVER, let alone doing it 20 times in a row to get into a slowly cleaner, uncluttered, more ready OS.

And then HOW they use their PC, an Optane wouldn't help them much, because the OS is only ever a gateway to them. Not most of the entire XP it is to ME installing them, let alone that I don't game much or use extraneous Apps much. My use is close to the OS. Their's is into a bubble cloud I've all built up for them.

But it is still their money, and I wouldn't especially advise the extra money into an OS-only optimized Optane drive.

 

[LabRat 891]

I just feel the need to say I entirely agree with this. A SN850X was by no way faster on an accordingly 4.0 platform. If anything, and if not especially with BOOT/REBOOT/WAKE UP, they were all slower. None of my customers or friends would see this. I will see this because I install Windows 6-8x a year either for myself or for them. I AM THE ONLY ONE that will see this, out of all of them.

And my friend that I've rebuilt, or his friend that I've built anew from far lesser specs and HDD reality, have NOTHING to complain about of course. Basically, they never need to reboot Windows 20x like I do when installing and refining their system. They don't neither have to wait through the bulk of a fresh OS install updates. Nor do they have to download and install most of the basic apps they use. I do all of this for them. Nor do they need to get rid of all the shit Win 10 and especially Win 11 will put onto them.

From there, what they DO NEED to do, and how often they DO NEED to reboot... is not something they can even feel or calculate. So then it's mostly waking up, and as much as old systems in DDR3/Sata SSD are a bit faster there, it's not like booting or rebooting. It has no REAL WORLD impact on them. They don't do most of the shit I do EVER, let alone doing it 20 times in a row to get into a slowly cleaner, uncluttered, more ready OS.

And then HOW they use their PC, an Optane wouldn't help them much, because the OS is only ever a gateway to them. Not most of the entire XP it is to ME installing them, let alone that I don't game much or use extraneous Apps much. My use is close to the OS. Their's is into a bubble cloud I've all built up for them.

But it is still their money, and I wouldn't especially advise the extra money into an OS-only optimized Optane drive.

Basically,
Optane was a 'miracle' but, one that is incredibly hard to 'sell' consumers and companies on.

Atop that, the "stacking latency" common to all modern systems, 'effectively' (read: apparently) neuters the benefits Optane brought.


Little to me is more irritating, than regularly re-realizing my 3rd gen i5 and 120GB SATAII SSD build that I use for Speedify VPN ISP bonding, is absolutely the 'snappiest' and quickest booting build in the entire house.

You're the first person in awhile I've seen even bring up the 'latency issues' inherit in modern systems...
(I can recall, about the time I got my X570 and R5 3600, pro gaming teams were switching back to Ivy Bridge. Because of latency.)

 

[Selaya]

it's not hard to sell to enterprise. the people who'd need something like optane, they know perfectly well what they've got w/ it.
the reason why it isn't being advertised much and why stocks are dwindling et cetera is bc the tech's tied up in patent struggles. not because it's unviable, or anything.

 

[LabRat 891]

it's not hard to sell to enterprise. the people who'd need something like optane, they know perfectly well what they've got w/ it.

At this point, the performance and endurance needed is being done with lots of cheap NAND.
-Years-back, before there were QLC Enterprise-rated drives, I'd agree.

the reason why it isn't being advertised much and why stocks are dwindling et cetera is bc the tech's tied up in patent struggles. not because it's unviable, or anything.

Correct.
However, I don't believe even resolving those issues would allow Optane to become competitive in today's age.
Excepting, Intel solidly keeping the IP, and immediately licensing it to anyone capable of fabricating it.

 

[bruhmoment01]

If anyone is interested in further benchmarks I’ve got a pair of Optane 905P drives, four P1600X drives, and four 512GB Optane DCPMM in my workstation.



Optane is much more interesting in the DIMM form factor and use case; this never made its way down from Xeon to consumers.

The CPU’s memory controller interleaves the DIMMs, and it can be configured either as storage or as volatile memory in the UEFI. You can even boot your OS from the DIMMs if it has support (RHEL or Fedora support this OOTB).

In memory mode, your DRAM is converted to an L4 cache layer before data reaches the Optane (this is the mode I use; 512GB DRAM + 2TB Optane).

random performance (the metric you'd care for) will actually get worse

I’ve tried out VROC RAID 0 on both the 905P and P1600X and the latency does indeed get much worse.

 

[QcRMoD]

Basically,
Optane was a 'miracle' but, one that is incredibly hard to 'sell' consumers and companies on.

Atop that, the "stacking latency" common to all modern systems, 'effectively' (read: apparently) neuters the benefits Optane brought.


Little to me is more irritating, than regularly re-realizing my 3rd gen i5 and 120GB SATAII SSD build that I use for Speedify VPN ISP bonding, is absolutely the 'snappiest' and quickest booting build in the entire house.

You're the first person in awhile I've seen even bring up the 'latency issues' inherit in modern systems...
(I can recall, about the time I got my X570 and R5 3600, pro gaming teams were switching back to Ivy Bridge. Because of latency.)

At this point, the performance and endurance needed is being done with lots of cheap NAND.
-Years-back, before there were QLC Enterprise-rated drives, I'd agree.

Correct.
However, I don't believe even resolving those issues would allow Optane to become competitive in today's age.
Excepting, Intel solidly keeping the IP, and immediately licensing it to anyone capable of fabricating it.

If anyone is interested in further benchmarks I’ve got a pair of Optane 905P drives, four P1600X drives, and four 512GB Optane DCPMM in my workstation.

View attachment 339930

Optane is much more interesting in the DIMM form factor and use case; this never made its way down from Xeon to consumers.

The CPU’s memory controller interleaves the DIMMs, and it can be configured either as storage or as volatile memory in the UEFI. You can even boot your OS from the DIMMs if it has support (RHEL or Fedora support this OOTB).

In memory mode, your DRAM is converted to an L4 cache layer before data reaches the Optane (this is the mode I use; 512GB DRAM + 2TB Optane).



I’ve tried out VROC RAID 0 on both the 905P and P1600X and the latency does indeed get much worse.

View attachment 339931

Incredible!

My dream comes true.

I thought I had posted all of this pretty much in vain. And could hardly come up first with my simple thought that nowadays latencies kill the idea.

Yet I remember vividly when DDR4 came forth to everyone… that basically the idea was out there. Latencies killing the potential of it.

So then can we have a real discussion about it… or will just everyone else hide behind the fact that their newer stuff is just better ONCE the OS is properly loaded into ever increasing needs for RAM/OS/nVME cache and now built in RAM etc.

My new core system and all others built into DDR5 platforms being even slower than DDR4 ones and none of the latter up to simple nVME 2.0 that did coexist alongside DDR3 and simpler CPU architectures?

Which made the best of the first Optane cache drives indeed that they came out as we were transitioning into ddr4 systems.

Which still makes sense today as we are very much burried under two layers of latencies since ddr5 arrival?

But which could not hold a candle to ddr3/sata/2.0 nVME systems…

Or that newer CPUs need to have a few cores and some 1.5GHz built in extra to just about be 50% faster than some good old 4/8 CPUs with no frills.

 

[Frozoken]

I doubt anyone will ever find this thread in Google search results, but a tl;dr for the above: if you're benchmarking your drive using fio (which KDiskMark uses as a backend) and want comparable results to CrystalDiskMark, make sure you have the option --scramble-buffers=0 and don't have the option --refill-buffers. The short explanation is that KDiskMark, by default, generates new random data for each block it writes, so in addition to benchmarking your drive, you're also benchmarking your CPU's ability to generate random numbers which... isn't really what you want.

A quick re-test after doing the above gives me a couple more reasonable results:

Sequential 1MiB Q1T1 write: 1017 MB/s
Sequential 1MiB Q1T1 mixed: 1230 MB/s
Random 4KiB Q1T1 write: 322 MB/s
Random 4KiB Q1T1 mixed: 388 MB/s

This only affects write workloads, so I didn't bother redoing the read tests.

Ive managed to get close to your "glitched" score except legitimately with heavy tuning tho so if ur dissapointed you can get that number genunely with enough work ahaha.

As a final to this thread, I'm looking at both the Optane 1600X and the 970 EVO when tested on my new core system.

So for all the other posts I was testing with a 7700K - M9C - 16GB 3200MHz C16/32GB 3000MHz C15 DDR4 (where the extra RAM borrowed from my now former server didn't change anything). All of these were ran obviously with the PCIe 3.0x4 capacity from that Z290 motherboard.

The results below were obtained on:

i5 14600KF (no OC - XMP Profile loaded)
Z790 PRO RS
32GB 6000MHz C30

Where the Optane was concerned, everything straightforward as usual, except of course we get some upscale on the new platform. Tests ran twice to ensure congruence, showing "best" result.

Left/Top = Z270 platform / Right/Bottom = Z790 platform


View attachment 339876View attachment 339879




View attachment 339875View attachment 339878





View attachment 339877View attachment 339880




Such an upscale is much less obvious however with the 970 EVO, while there was still a trend with RND4K getting weird results, however it seems to have somewhat shifted in the Mix instead of the Reads for most runs EXCEPT in RND4K Q1T1 where I COULD NOT regularize Reads/Mix at all neither reciprocate Mix especially: either Reads would fail, or sometimes both failed, and it shifted like that on all (nearly a dozen) runs I did. So I kept two results there to image it.

View attachment 339882View attachment 339885





View attachment 339881View attachment 339884





View attachment 339883View attachment 339886


View attachment 339887




So the Def Peak run showed a slight improvement in RND4K Q8T1 with the 970 EVO, although it's not very substantial. For the rest of the tests, it's all pretty much flatlining.


For an extra effort, I compared both drives as well in ATTO, using the "per default" test, then changing the options until I found the ATTO test that was the most "excruciating" for both drives, which is to Bypass Write Cache, adding Verify Data, which in turns gives the option to change the test pattern to "Random" (per default would be "00000000" which seems to be the same thing than 0Fill in CDM) and to change the Queue Depth to "1" instead of "4".

Left/Top is the Optane, Right/Bottom is the 970 EVO.

View attachment 339893View attachment 339891


View attachment 339892View attachment 339890

Where Optane is superior until 8K blocks, both drives about equalizing at 16K, and the 970 EVO being superior from 32K onwards.

And THAT'S IT. No more testing, no more discussion of testing, I am done with adding new data to this thread.

Thanks all for patience and participation.

Your RND 4k results improve with the better cpu because its so fast that your cpu with one thread cant fully keep up and bottlenecks its latency partially. This is why me as someone with a heavily overlocked 13600k including ram, ring, power saving features off etc has gotten the highest result ive seen on it becuase i have the fastest cpu out of anyone testing. C states off though alone gave me like +20MB/s qd1 random reads because theres definitely some wake up latency bottlenecking how quick your optane can respond yet again. Nand devices on the other hand are like 5x slowr minimum so they arent an issue for reasonable systems

Nothing wrong indeed. Optane had the future of SSDs right.

But now it's exctinct and beyond capabilities in RND4K Q1T1, what really is there? Outlandish $/GB is there. NO ONE CARES because it's not something readily felt or needed towards capacity and $/TB.

And that IS because OS and primary apps are a gateway into something that will just NOT USE an Optane capabilities - well, at least, not nearly as much as synthetic tests would show.

Let me just give everyone an example and propose that they shall prove me wrong:

I've just transfered my best friend from an i5 4570 DDR3 on an old HyperX Fury Sata 120GB drive to a 12600K/SN850X. Then I've restarted his old PC in Win 11 still on that old POS Sata SSD 120GB.

I've built a friend of my friend a 13600K/Z790 PC DDR5 a year ago with a SN770.

I've reinstalled my old PC on an Optane 1600X in Win 10 and Win 11 compared to doing it with an old 970 EVO.

I've then upgrade my old PC with Z790 i5 14600K just the last week on Win 11.

Guess which system is BY FAR THE FASTEST TO EITHER BOOT, REBOOT, OR WAKE FROM SLEEP?

It's the 4670 DDR3 with the old Sata SSD 120GB... and I'm not talking quantum differences... I'm talking HUGELY speedier.

DDR4 and DDR5 and newer CPUs if not RAM.... they've so much latency everywhere even with Optane 1600X... they FEEL SLOW and ARE INDEED SLOW in all of those three operations that will mostly define Everyone feeling of a speedy OS. By FAR.... it's like 250% faster on a Sata SSD and DDR3.

And then what we're talking about with Optane is faster background/launching apps benefits? And as rebooting/booting/waking up is concerned, an old DDR/SSD Sata feels speedier.

Yeah do not convince me.... I got the Optane.

Just convince me to convince ANYONE OF MY REGULAR CUSTOMERS, let alone FRIENDS, that they should purchase into it.

True altho that is due to the extra ddr4 and especially ddr5, initialisation and training time it has to do every boot, NOT latency.

 

[QcRMoD]

Ive managed to get close to your "glitched" score except legitimately with heavy tuning tho so if ur dissapointed you can get that number genunely with enough work ahaha.


Your RND 4k results improve with the better cpu because its so fast that your cpu with one thread cant fully keep up and bottlenecks its latency partially. This is why me as someone with a heavily overlocked 13600k including ram, ring, power saving features off etc has gotten the highest result ive seen on it becuase i have the fastest cpu out of anyone testing. C states off though alone gave me like +20MB/s qd1 random reads because theres definitely some wake up latency bottlenecking how quick your optane can respond yet again. Nand devices on the other hand are like 5x slowr minimum so they arent an issue for reasonable systems


True altho that is due to the extra ddr4 and especially ddr5, initialisation and training time it has to do every boot, NOT latency.

Interesting I did not retest with my modest OC and I do have C states enabled. Which will remain this way because it’s realistic of my use, but I’ll retest just to see about the OC.

And well yeah I agree about latency in the sense that I was not especially talking about CAS latency but just « latency » as a generic term but it is probably better to be precise and I appreciate your pinpointing the proper distinction.

 

[Frozoken]

And well yeah I agree about latency in the sense that I was not especially talking about CAS latency but just « latency » as a generic term but it is probably better to be precise and I appreciate your pinpointing the proper distinction

Nah that's my bad it made sense in your context now that I think about it lol. Oh yeah on the cpu bind thing ive realised under a rnd 4k qd1 load on crystal disk mark the drive is only at 70% use lmfao. If that means it can scale to what that implies being ~470MB/s/0.7 it'd mean we get like 670MB/s qd1 reads from the drive with better cpus

 

[QcRMoD]

Nah that's my bad it made sense in your context now that I think about it lol. Oh yeah on the cpu bind thing ive realised under a rnd 4k qd1 load on crystal disk mark the drive is only at 70% use lmfao. If that means it can scale to what that implies being ~470MB/s/0.7 it'd mean we get like 670MB/s qd1 reads from the drive with better cpus

No bad there, on a PC enthusiast forum, probably most people would rather think you made a proper distinction, whereas I was flaunting "latency" a bit too liberally. Anyway, no biggie one way or another.

Update on Optane results with my OC... and while I was there, I also retested the 970 EVO see if the same kind of scaling applied.



In the SEQ128K Q32T1 nVME Peak tests, both drives tested about the same they did without the OC on all fronts (with 5.4 GHz OC on the left, without on the right and same as posted last month):

In Default Peak, both drives get a good push forward with RND4K Q32T1:

Bizarely, the 970 EVO gets a "funny" push where Writes somehow almost got to the same speed than Reads. Of course, I re-ran the 970 EVO a couple of times because its behavior is still a bit erratic as always, and the above was "confirmed" but I'm suspecting the Reads should be faster theoretically, yet I could not manage to make it happen, and you know, maybe it's what it should be.

So in the Optane's case, we get roughly +35% in Reads and +50% in both Writes and Mix. With the 970 EVO, we get roughly +15% on Reads, +50% on Writes, and +30% on Mix.


In RND4K Q1T1, a smaller push, but still tangible.

Very linear scaling on both sides: roughly +15% on all Reads Writes and Mix here. Well, arguably the 970 EVO gets more of a +10% bump only in Mix, but at those poor speeds anyway, nothing as impressive as with the Optane there.

Thanks @Frozoken for bringing this to my attention.

 

[Frozoken]

No bad there, on a PC enthusiast forum, probably most people would rather think you made a proper distinction, whereas I was flaunting "latency" a bit too liberally. Anyway, no biggie one way or another.

Update on Optane results with my OC... and while I was there, I also retested the 970 EVO see if the same kind of scaling applied.



In the SEQ128K Q32T1 nVME Peak tests, both drives tested about the same they did without the OC on all fronts (with 5.4 GHz OC on the left, without on the right and same as posted last month):

View attachment 342895View attachment 342910

View attachment 342892View attachment 342909




In Default Peak, both drives get a good push forward with RND4K Q32T1:

View attachment 342894View attachment 342899

View attachment 342891View attachment 342897

Bizarely, the 970 EVO gets a "funny" push where Writes somehow almost got to the same speed than Reads. Of course, I re-ran the 970 EVO a couple of times because its behavior is still a bit erratic as always, and the above was "confirmed" but I'm suspecting the Reads should be faster theoretically, yet I could not manage to make it happen, and you know, maybe it's what it should be.

So in the Optane's case, we get roughly +35% in Reads and +50% in both Writes and Mix. With the 970 EVO, we get roughly +15% on Reads, +50% on Writes, and +30% on Mix.


In RND4K Q1T1, a smaller push, but still tangible.

View attachment 342896View attachment 342900

View attachment 342893View attachment 342898

Very linear scaling on both sides: roughly +15% on all Reads Writes and Mix here. Well, arguably the 970 EVO gets more of a +10% bump only in Mix, but at those poor speeds anyway, nothing as impressive as with the Optane there.

Thanks @Frozoken for bringing this to my attention.

Very interesting results. Was your OC consisting of power saving features being turned off too because those should benefit both whole clock speed should only affect optane + ssd writes and you only ocd 100mhz there. Also yeah ssd writes improve because it's basically all just ram cache and ofc like optane, ram is fast enough to be bottlenecked by the cpu.

Why are writes basically all ram? Because unlike optane, nand has ridiculously asymmetrical.write speeds naturally and u can test by turning off your cache in windows. In my 860 evo, random.writes dropped from 200MB/s to 2MB/s with cache off . Reads were unaffected as (idk if people are aware of this) ssd dram only is boosting writes.

Either way +50% is wild i think something else is going on too lol because damn. Either that ir speedshift is that shit bcuz I never use it. Oh yeah also on both obviously with enough queues at 1 thread youre going to be completely cpu bound but q32 doesn't happen in the real world even if it's enough to get even nand held back by the cpu.


Ill do a follow up with just changing my clock speeds on the same install and hardware as a control

No bad there, on a PC enthusiast forum, probably most people would rather think you made a proper distinction, whereas I was flaunting "latency" a bit too liberally. Anyway, no biggie one way or another.

Update on Optane results with my OC... and while I was there, I also retested the 970 EVO see if the same kind of scaling applied.



In the SEQ128K Q32T1 nVME Peak tests, both drives tested about the same they did without the OC on all fronts (with 5.4 GHz OC on the left, without on the right and same as posted last month):

View attachment 342895View attachment 342910

View attachment 342892View attachment 342909




In Default Peak, both drives get a good push forward with RND4K Q32T1:

View attachment 342894View attachment 342899

View attachment 342891View attachment 342897

Bizarely, the 970 EVO gets a "funny" push where Writes somehow almost got to the same speed than Reads. Of course, I re-ran the 970 EVO a couple of times because its behavior is still a bit erratic as always, and the above was "confirmed" but I'm suspecting the Reads should be faster theoretically, yet I could not manage to make it happen, and you know, maybe it's what it should be.

So in the Optane's case, we get roughly +35% in Reads and +50% in both Writes and Mix. With the 970 EVO, we get roughly +15% on Reads, +50% on Writes, and +30% on Mix.


In RND4K Q1T1, a smaller push, but still tangible.

View attachment 342896View attachment 342900

View attachment 342893View attachment 342898

Very linear scaling on both sides: roughly +15% on all Reads Writes and Mix here. Well, arguably the 970 EVO gets more of a +10% bump only in Mix, but at those poor speeds anyway, nothing as impressive as with the Optane there.

Thanks @Frozoken for bringing this to my attention.

UPDATE:
I dropped my clockspeeds from 5.6ghz to 3ghz all core and in 4k qd1(MB/s) my optane dropped from 470/360 to 335/265. My nvme wd sn850x on the other hand dropped from 83/380 to 80/290 so much less read scaling as suspected.

 

[QcRMoD]

Very interesting results. Was your OC consisting of power saving features being turned off too because those should benefit both whole clock speed should only affect optane + ssd writes and you only ocd 100mhz there. Also yeah ssd writes improve because it's basically all just ram cache and ofc like optane, ram is fast enough to be bottlenecked by the cpu.

Why are writes basically all ram? Because unlike optane, nand has ridiculously asymmetrical.write speeds naturally and u can test by turning off your cache in windows. In my 860 evo, random.writes dropped from 200MB/s to 2MB/s with cache off . Reads were unaffected as (idk if people are aware of this) ssd dram only is boosting writes.

Either way +50% is wild i think something else is going on too lol because damn. Either that ir speedshift is that shit bcuz I never use it. Oh yeah also on both obviously with enough queues at 1 thread youre going to be completely cpu bound but q32 doesn't happen in the real world even if it's enough to get even nand held back by the cpu.


Ill do a follow up with just changing my clock speeds on the same install and hardware as a control


UPDATE:
I dropped my clockspeeds from 5.6ghz to 3ghz all core and in 4k qd1(MB/s) my optane dropped from 470/360 to 335/265. My nvme wd sn850x on the other hand dropped from 83/380 to 80/290 so much less read scaling as suspected.

I’d like to give a real effort in answering as I’m sure you did, but I’m still not sure what it is I could say.

Honestly it is what it is. At some point in the early I’ve realized my motherboard had a stupid setting called ASRock DRAM Frequency Optimization that would override XMP profile and run some base JEDEC like 5200 MHz CL44 or something instead of 6000MHz CL30. It was probably AFTER publishing here because I can’t remember realizing it and putting it right when I was still busy on setting Win 11 and indeed testing the Optane. Yet I cannot remember not too neither.

If something like 800MHz more RAM speed
to about 25% better latency + 100MHz on each Pcores + 100 MHz on each E-Core and Ring fixed to 47 with ring to ratio offset disabled can help explain it, then yeah probably it’s just all of that and all apologies for not putting figures right.

As already said however I do not disable any C-States ever, and I do run a rather strict Power Plan at all times. I’ve never seen a point not to save Power or let things go to sleep. But I know of C-States throttling and any time I bench anything I push the Power Plan back to a very slack 1/2 hours screen/sleep just to make sure it doesn’t directly enter the equation of 5/30 minutes bench tests. I think for the Optane I just used Never, but it could have been just the 1/2 hours push back.

Hope this can clarify the results - still very average - and of anything more that might have went into there.

Realizing the stupid MB DRAM override setting, to logical extents I’d consider truest, would have come after posting here last month… most very probably. I’d like to be sure but I can’t.

In fact I do clearly remember remarking it in Task Manager, overriding it in DRAM clocks first, and even running the EXPO profile instead for a while cause it would override the override. Then I realized that stupid setting was not about enhancing POST or training as I first thought, but just getting in the way where it should not. But I cannot still give all of that any timeline except « in the early » and if I posted about RAM being at 6000MHz well I verified it first. Probably the cause for overriding then running the EXPO profile instead to get there.

 

[RJARRRPCGP]

Looking at these benches, because I think I should focus on random-access-speed, looks like I'm better off RAID-0'ing SATA SSDs! I have a bunch of them!

Rather then trying to RAID-0 NVMe! I don't know if I can even pull that one off well with less than X570!

 

[dgianstefani]

Every drive you add in RAID 0 Stripe increases the chance of failure.

Just grab an Optane as boot media and be done with it.

 

[Frozoken]

I thank you very much for your interest and participation.



Seeing how it CAN deal with as little 512B segments, this is probably expected. ATTO SSD benchmark can be interesting in that regard.

View attachment 337372View attachment 337371

But asking synth benchmarks to tell the whole story is probably useless. Also I do lack comparisons with ATTO of the SN850X or the 970 Evo.

I'd like to see one proper review/expansive sets of tests carried on it. But I guess what I did is the best we'll get out of it until someone with the tools and knowledge to really say/show some things about it will arise.

Late but here's my atto which I think is maxing out my cpu on read because it went from 70MB/s at 512B to nearly 130 when forcing it on the Pcores (it was defaulting to using the ecores, thanks intel).

 

[Shrek]

Finally found this thread on the Optane 1600X 118GB, tempted to get one as a boot drive.

 

[dgianstefani]

Finally found this thread on the Optane 1600X 118GB, tempted to get one as a boot drive.

View attachment 351583

Do it.

No regrets here.

Big NVMe as secondary bulk storage. No real need for a third internal drive as of yet.

 

[Shrek]

Seems things are still very much supported.

Intel® Optane™ SSD P1600X Series

I wonder about your latency numbers, since Intel doesn't provide latency standards for the P1600X, but for other Optane drives it's usually supposed to be around or less than 10 µs.

• 7 μS Read
• 10 μS Write

 

[Frozoken]

Seems things are still very much supported.

Intel® Optane™ SSD P1600X Series

• 7 μS Read
• 10 μS Write

View attachment 351624

It is i dont even see how an ssd becomes "unsupported". As long as the generic microsoft nvme driver keeps getting updated (obv it will), itll be "supported"

 

[Shrek]

I meant the unit is supported for firmware, not drivers.

 

[dgianstefani]

It is i dont even see how an ssd becomes "unsupported". As long as the generic microsoft nvme driver keeps getting updated (obv it will), itll be "supported"

No it's a "Windows Vista ready keyboard!"

 

[Shrek]

Finally got to install the 118GB Optane as a boot drive (running Windows 11)

I move game downloads onto a separate drive, but things like MS Office seem to install by default on the boot drive so I didn't fight that and I leave my own work on the desktop, as a result my 118GB Optane drive sits at almost exactly 50% full.

I copy a backup of my work onto the separate drive, so I don't worry about the boot drive failing.