Thanks. Not sure how to do that, and I think this title is fine anyway.
I haven’t seen anywhere that documents the meaning of each exact POST code (nor the order of the bits - though perhaps I’m overthinking it). Perhaps the codes have different meanings for each revision of the board? Or I’m just missing something.
I’m not an expert - just parroting information I’ve found that might be topical. Doubt you could really cause “harm” though.
I’d say having let it sit for 45 minutes is good enough to say it’s not going to work no matter how long you wait. Trying another kit is definitely a good idea to decide bad laptop vs incompatible ram, since I doubt you could immediately be getting replacement boards when they’re still in this early production period.
You were right… The error code for no RAM is different.
Edit: I don’t know why Amazon is being so cool about it, but they are apparently going to allow me to return the kit. I’ll lose 20% for a restocking fee, but at least I won’t have to deal with selling them on eBay. Could be defective for all I know.
Edit 2: @Justin_Stackhouse, did you try killing all power to the mainboard (including the battery) for a little while? Didn’t occur to me until a few minutes ago. Doesn’t seem to have fixed the issue for me, but I would expect yours to work… Sorry that its not!
No luck on anything I tried. I am hopeful that the issue is the RAM, I’ll be near a micro center on Saturday visiting family so I’ll be able to try different options to get up and running.
I will say that I wish there were more public information on the POST codes in Framework’s documentation, but I guessing there are some restrictions from partners on this. Here’s hoping I get a quick resolution from support, ideally tomorrow, so I can enjoy this wonderful device.
I just mean that Kingston Fury Impact and I think one of the G Skill kits have more aggressive default timings than the standard JEDEC profile that they will attempt to boot into even without XMP enabled.
That table looks nice, but isn’t completely accurate.
For example at 5600 MT/s that table indicates that the JEDEC classes are: A (40-40-40), B (46-46-46), and C (50-50-50).
Whereas this massive document from Micron states that the JEDEC classes at 5600 MT/s are: AN (40-40-40), B (46-45-45), BN (46-46-46), and C (50-49-49).
We can confirm that the Anandtech table is inaccurate because kits like the Crucial CT2K16G56C46S5 exist with 46-45-45 JEDEC timings, which wouldn’t be possible with the Anandtech table.
Weird, anadtech lists the literal jdec specification as a source but I am not sure how exactly those bins work (and I am not paying 400 bucks for the document XD), it may well be lower limits so anything between has to work for the memory controller anyway and the actual values are read from the spd.
The micron values are all still within the min/max from the anadtech JDEC values so I’d guess both are accurate. Micron B and C bins are just slightly better than absolutely required but I am just guessing at this point.
5600 CL40 is still a valid spec though be it from anandtecs interpretation of the jdec doc or from this micron datasheet.
I’ve had some back and forth with support that is continuing. My issue is an ongoing comedy of errors.
Firstly, 1 out of 2 of my Crucial CT2K16G56C46S5 is almost certainly bad, not working in any slot or combination.
Second, one of my issues is that the Memory Slot 0 QR code was askew and partially covering some pins on the MB. I only noticed this while getting pictures of the pins under the flaps to send for them to investigate. Very frustrating, and compounded with the RAM stick (not from Framework) being bad made it a nightmare.
For now I have a single channel memory laptop that is up and running and I’ll take it as a win.