Switch to full style
CompactFlash, SD, MMC, USB flash storage. Anything that does not have moving parts inside.
Post a reply

Microsoft Surface 1807 strange chip

July 13th, 2018, 23:12

Hi everyone!
Have anyone seen this chip?
Is there any kind of interface for that?
Attachments
IMG_2628.JPG

Re: Microsoft Surface 1807 strange chip

July 13th, 2018, 23:30

It is strange the chip part number is the same as Intel SSD Part Numbers. maybe a fully integrated drive on a chip, a monolith SSD essentially?

Re: Microsoft Surface 1807 strange chip

July 14th, 2018, 2:00

"A fully integrated drive on a chip" sounds to be right.
What on Earth is happening in a flash world?!?!

Re: Microsoft Surface 1807 strange chip

July 14th, 2018, 3:00

Ronald111 wrote:"A fully integrated drive on a chip" sounds to be right.
What on Earth is happening in a flash world?!?!


Making our life difficult... :wink:

PM sent...

Re: Microsoft Surface 1807 strange chip

July 14th, 2018, 5:00

Here is a SATA/PATA "SSD on a chip" (Silicon Motion FerriSSD):

http://static6.arrow.com/aropdfconversion/e34a936206aef41642e4466404978f1b53963ba5/ferrissdproductbrief_eng.pdf
http://www.storagesearch.com/ferrissd.pdf

Toshiba and Samsung both have NVMe SSD-on-a-chip products.

I suspect that the vacant BGA location above the Intel chip is reserved for a second SSD. This should make it easier to determine the pinning for the supply rails and interface signals.

Re: Microsoft Surface 1807 strange chip

July 15th, 2018, 8:39

Single-package SSDs fall under the PCI-SIG, and there are four standard form factors. The Toshiba and Samsung products use the Type 1620 package that measures 16mm by 20mm. The form factor can operate over a SATA 6Gb/s bus or PCIe 3.0 with either two or four lanes (presumably NVMe).

https://www.tomshardware.com/news/intel-bga-nvme-ssd-600p,36172.html
1620.png



pinout (Arguably useful) : in this doc: https://pcisig.com/sites/default/files/specification_documents/BGA_SSD_ECN.pdf

pinout.png


Even this may be old hat even before they admit it exists. want something 1,000 faster than current SSD? [urlhttps://www.geek.com/chips/new-intel-storage-is-1000-times-faster-than-your-ssd-1629656/[/url]

Re: Microsoft Surface 1807 strange chip

July 15th, 2018, 12:48

Yeah, I had one of these show up already. Fortunately I was able to boot the machine to Linux from a USB and image the SSD out to an external HDD.

Re: Microsoft Surface 1807 strange chip

July 15th, 2018, 18:18

I wonder if the future of solid state storage is socket SSDs.

Re: Microsoft Surface 1807 strange chip

July 15th, 2018, 18:22

fzabkar wrote:I wonder if the future of solid state storage is socket SSDs.


No, I think the future is more and more shifting toward single board and monolith design. The less opportunities people have to repair a product, the more often they'll need to buy a new one. The days of upgradable computers is dying out my friend. Someday we'll all go the way of the TV repair man. But, who knows, maybe by then we'll all be repairing space-x shuttles or something else complex that's prone to breaking all the time.

Re: Microsoft Surface 1807 strange chip

July 15th, 2018, 19:29

I don't think people sit around and actually decide to make something less repairable, like as a business decision. a socket costs, and sockets are used only when necessary. BGA does away with a small cost. I do think however that the same is true that they also are not factoring in repair at all. They are simply trying to increase storage/speed and decrease cost.

an upgradable BGA socket for one of these chip SSDs is a great idea though. Many motherboards are coming out with M2 slots, so I think Franc is pretty close in his prediction, but I wonder how fast that too would go out of fashion.

Amazing how little information is out there on these chip SSDs - you would think it would be a product to be proud of and advertise. I guess device manufacturers have a more direct line to storage vendors, and this kind product is fine kept under NDAs and such.

All I know is Flash recovery is going to continue to be a complete pain in the ass.

Re: Microsoft Surface 1807 strange chip

July 15th, 2018, 21:30

And another one came today.
Now it is Surface Laptop.
Desperately need some socket solution.
Attachments
IMG_2633.JPG
2018-07-16_10-25-13.png

Re: Microsoft Surface 1807 strange chip

July 16th, 2018, 1:34

You can make one. !!
https://www.google.com.au/amp/s/m.aliexpress.com/item/32823128771.html

Re: Microsoft Surface 1807 strange chip

August 2nd, 2018, 12:04

data-medics wrote:Yeah, I had one of these show up already. Fortunately I was able to boot the machine to Linux from a USB and image the SSD out to an external HDD.

Unfortunately mine will not boot into recovery or via a USB (Windows or Linux).

Re: Microsoft Surface 1807 strange chip

August 2nd, 2018, 13:55

Make sure you're using the right UEFI and Secure Boot settings in the BIOS. It took me some figuring out to get a linux USB to boot and not all builds would work, some just crashed.

Re: Microsoft Surface 1807 strange chip

August 2nd, 2018, 14:09

It wouldn't be too bad to just wire it up like a monolith I suppose. I doubt it'd require power input on all the different voltages either. My guess is it's just different options the manufacturer can opt to use to power it. I'd guess that just using the 3.3 or 1.8 would suffice.
Attachments
Toshiba SSD Connection Points.jpg
Toshiba SSD Connection Points.jpg (75.53 KiB) Viewed 30008 times

Re: Microsoft Surface 1807 strange chip

August 2nd, 2018, 14:43

Should work if the controller is still functional I guess. If I can't see anything significant on the board and decide to take the chip off I will let you know. In between the two posts I did work out the secure boot issue and got it to boot into Linux and I can see the SSD which is positive.

By the way, how did you image the drive after booting from Linux when it only has one USB port?

Re: Microsoft Surface 1807 strange chip

August 2nd, 2018, 14:50

ddrecovery wrote:By the way, how did you image the drive after booting from Linux when it only has one USB port?


Usb hub? :-)

Re: Microsoft Surface 1807 strange chip

August 2nd, 2018, 14:52

pcimage wrote:
ddrecovery wrote:By the way, how did you image the drive after booting from Linux when it only has one USB port?


Usb hub? :-)

Wasn't sure if booting from a USB on a hub and trying to save over the same hub would work. I guess it does. Many thanks :-)

Re: Microsoft Surface 1807 strange chip

August 2nd, 2018, 15:06

ddrecovery wrote:
pcimage wrote:
ddrecovery wrote:By the way, how did you image the drive after booting from Linux when it only has one USB port?


Usb hub? :-)

Wasn't sure if booting from a USB on a hub and trying to save over the same hub would work. I guess it does. Many thanks :-)


Yep, I just used a self-powered hub and plugged in both the USB to boot it and a dongle for wireless keyboard/mouse. Then after booting I plugged in an external hard drive.

Worked like a charm and got a full ddrescue image in my case.

I found that the WiFi was very unreliable in linux for me. Not that you need the WiFi, I was just trying to install something on my live USB that wasn't already there, but it'd drop offline and never work again until I'd reboot. I'm curious if you have the same issue.

Re: Microsoft Surface 1807 strange chip

August 3rd, 2018, 19:08

I am using Ubuntu 16.04 from a live USB but am having difficulty installing ddrescue on the USB. I can use R-Studio to image but the NAND is pretty unstable and keeps dropping out at about 8% done. I ham oping to image backwards and see if that makes a difference. I have tried installing ddrescue on the live USB in another system and get the same issue so it not the Surface WiFi causing the problem. Is there something special that needs to be installed in order to get ddrescue to work on a live USB?

I have also tried DEFT Zero as it has ddrescue built in but it will not boot on this Surface.
Post a reply