At Computex 2018, Intel announced a new Optane 905P SSD:
…the Optane 905P 380GB, now in an M.2 form factor!
This looks to be a miniaturization of the 7-channel controller previously only available on the desktop add-in cards (note there are 7 packages). There is a catch though, as fitting 7 packages plus a relatively large controller means this is not M.2 2280, but M.2 22110. The M.2 22110 (110mm long) form factor may limit where you can install this product, as mobile platforms and some desktop motherboards only support up to an M.2 2280 (80mm) length. Power consumption may also be a concern for mobile applications, as this looks to be the full blown 7-channel controller present on the desktop AIC variants of the 905P and 900P.
We have no performance numbers just yet, but based on the above we should see figures in-line with the desktop Optane parts (and higher than the previous 'Optane Memory'/800P M.2 parts, which used a controller with fewer channels). Things may be slightly slower since this part would be limited to a ~7W power envelope – that is the maximum you can get out of an M.2 port without damaging the motherboard or overheating the smaller surface area of an M.2 form factor.
An interesting point to bring up is that while 3D XPoint does not need to be overprovisioned like NAND flash does, there is a need to have some spare area as well as space for the translation layer (used for wear leveling – still a requirement for 3D XPoint as it must be managed to some degree). In the past, we've noted that smaller capacities of a given line will see slightly less of a proportion of available space when comparing the raw media present to the available capacity. Let's see how this (theoretically) works out for the new 905P:
- 800P 58GB – 64GB RAW – 10%
- 800P 118GB – 128GB RAW – 8%
- 900P 280GB – 336GB RAW – 20%
- 905P 380GB – 448GB RAW – 18%
- 900P 480GB – 560GB RAW – 17%
- 905P 960GB – 1120GB RAW – 17%
I'm making an educated guess that the new 380GB part contains 4 die stacks within its packages. We've never seen 8 die stacks come out of Intel, and there is little reason to believe any would be used in this product based on the available capacity. Note that higher capacities run at ~17% excess media, but as the capacity reduces, the percentage excess increases. The 280GB 900P increases to 20% by that capacity, but the new 905P M.2 comes in at 18%. Not much of a loss there, meaning the cost/GB *should* come in-line with the pricing of the 480GB 900P, which should put the 905P 380GB right at a $450-$500 price point.
The new 905P M.2 22110 is due out later this year.
> no performance numbers
> no performance numbers just yet
Is the interface now x4 PCIe lanes on this latest Optane M.2?
Many thanks to you and Ken
Many thanks to you and Ken for staying close to recent developments in this area of high-performance storage.
p.s. Is there any way we can be designated PCPER-Fan-Men?
… maybe Fan-Persons? LOL!
yes it is x4 pcie lanes, but
yes it is x4 pcie lanes, but concern is this is 3.3v? cause most m.2 lots on motherboard or laptop are 3.3v
Yes, 3.3V is the concern
Yes, 3.3V is the concern (more specifically, the maximum power that can be safely delivered at that voltage).
does this imply that their
does this imply that their new controller on this 905p m.2 version is a different one. it must be if it uses less voltage and less power.
It would have to be since the
It would have to be since the 905P draws as much power at idle as an M.2 part would be limited to as a maximum. The controller should work the same (same channels, etc), but just be a lower power part. Intel did a similar thing when moving from the Optane Memory (16/32GB) parts to the newer 'M10' version of the same, as well as the 800P. Those had a nearly identical controller/performance yet used a fraction of the power, so I don't see why Intel couldn't pull off something similar with an M.2 version of the 905P.
thats very reassuring seeing
thats very reassuring seeing possibility having this to work off a 3.3v rail. imho the m.2 spec would be the same anywhere so intel shouldn’t make a m.2 device that has 12v rail into it.
The ASRock Ultra Quad M.2
The ASRock Ultra Quad M.2 “4×4” AIC has a 6-pin power input connector, same as video cards, which appears to be good advance planning.
If could find one for sale
If could find one for sale I’d love one for x299 i9 7980Xe sysyem
Intel separately reached out to us and said:
“We’ll also provide more details when the product becomes available later this summer.”
It seems like the availability information Intel has provided is conflicting, but we expect to see the drives released in the coming few months.
Concerning heat and power
Concerning heat and power demand:
However, even with a heatspreader over the SSD controller, it is likely that sustained workloads will lead to thermal throttling unless the 905P is housed in a more substantial cooling system, such as the actively-cooled quad-M.2 adapters.
Such throttling is also difficult to avoid for flash-based M.2 SSDs, but in general consumer storage workloads don’t provide enough sustained I/O for throttling to be a concern.
Optane SSDs are marketed precisely for the workloads that are an exception to that rule, so throttling will be harder to overlook.
So bought a apex x299 i9
So bought a apex x299 i9 7989xe system thinking id build this beast as time passed i got 64gb of ram, a vega Frontier GPU, then came storage i think i went overboard a lil overboard when finally got to upgrade from a 960 evo 256gb.
1.intel 900p 280gb u.2/pcie Adapter(Cache)
2.intel 750 1.2TB pcie (Steam)
3.intel 750 400gb pcie (Primo Cache)
4.intel 905p 380gb (Boot)
5.intel 800p 118gb (HDD Cache)
6.intel 3.84 TB SSD (Bulk Steam Games)
7.2x WD Black 2TB (Video Editing/Scratch)
8.WD 6TB Black 256mb cache HDD (System Backup)