huh, interesting. So you're saying we might also get 6GiB and 12GiB modules? Because 12GiB, and 24GiB configuration on laptops were really annoying, knowing you only had it working in single channel mode.
That's not exactly true. With 6GB (4GB + 2GB modules) you have 4 GB (2GB + 2GB) in dual channel and only the rest of the bigger module is in single channel. Actually, a 6GB module will make it possible to have 6GB in single channel mode :)
More modules more money. Laptop vendors are saving money with these setups. Also, asymmetric channel loading has never been in a single supported memory configuration of any motherboard or CPU. It might work, but I'd be shocked if it worked in dual channel mode.
A lot of modern Intel chips support it, mainly mobile chips IIRC. I can't recall seeing support for async dual channel on any AMD chips, which wouldn't really be an issue if laptop manufacturers didn't do weird s*%^ - like solder 4GB on the mainboard and leave a single DIMM slot to handle the other channel. Even if it was properly balanced to start with in a shipping configuration, it makes RAM "upgrades" really awful.
Oh that reminds me, I did have a Socket 7 system where you couldn't exceed 64MB of memory or you'd lose cache. Chipset limitation... 430TX or maybe older, can't recall now.
This is interesting. While it's true that cell phones and other mobile devices have been using some non-power-of-two memory devices for a while 24GiB devices to stay under the 4GiB limit of some 32bit SoCs, such sizes have never commonly been used on PCs. I'm very concerned with bugs in firmware and in chipsets supporting these sizes well there.
Unlike embedded systems where this type of memory has been used, PCs aren't fixed hardware with hardcoded memory sizes (and speed, timings, etc.). Given how many bugs we find in higher level PC software when past assumptions are challenged, I truely fear the many likely latent bugs we're going to find when NPoT memory sizes come to the PC.
I'd sort of feel safer if it came to the server side first, because it's likely to be actually tested there.
I suspect that as long as the hardware and OS can handle it, the software will be fine. VMs can be assigned non power-of-two ram amounts today. It's something I've made use of myself. When you've got an untold number of VMs all competing for a host's resources, they each get what RAM they need, not what makes their RAM pool a power of two.
I don't worry about the OS once it gets running. With three bus processors like the older Intel processors (Westmere, etc.), I have no concern with the OS dealing with arbitrary sizes, but the firmware that has to program all the address decoders for the chip select lines, etc. Those I worry about.
Any word on what part of the address is "absorbing" the non-power-of-two? I'd have to guess it's an odd number of bank groups, but who knows what kinds of shenanigans they may be up to.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
22 Comments
Back to Article
Adam7288 - Tuesday, December 14, 2021 - link
They should just call it 10drballsystemlord - Tuesday, December 14, 2021 - link
I want 5!!! (Not a power of two. ;) )StevoLincolnite - Wednesday, December 15, 2021 - link
12GB is still powers of two though. 8 bits in a byte still a power of 2 at the end of the day...Wrs - Wednesday, December 15, 2021 - link
Nah there's a 3 in there that prevents it from being 2^n where n is a whole numbermeacupla - Tuesday, December 14, 2021 - link
huh, interesting.So you're saying we might also get 6GiB and 12GiB modules?
Because 12GiB, and 24GiB configuration on laptops were really annoying, knowing you only had it working in single channel mode.
kobblestown - Wednesday, December 15, 2021 - link
That's not exactly true. With 6GB (4GB + 2GB modules) you have 4 GB (2GB + 2GB) in dual channel and only the rest of the bigger module is in single channel. Actually, a 6GB module will make it possible to have 6GB in single channel mode :)willis936 - Wednesday, December 15, 2021 - link
More modules more money. Laptop vendors are saving money with these setups. Also, asymmetric channel loading has never been in a single supported memory configuration of any motherboard or CPU. It might work, but I'd be shocked if it worked in dual channel mode.Darkknight512 - Tuesday, December 28, 2021 - link
This isn't correct at all, multiple platforms support asymmetric dual channel configs.Alexvrb - Sunday, January 2, 2022 - link
A lot of modern Intel chips support it, mainly mobile chips IIRC. I can't recall seeing support for async dual channel on any AMD chips, which wouldn't really be an issue if laptop manufacturers didn't do weird s*%^ - like solder 4GB on the mainboard and leave a single DIMM slot to handle the other channel. Even if it was properly balanced to start with in a shipping configuration, it makes RAM "upgrades" really awful.Oh that reminds me, I did have a Socket 7 system where you couldn't exceed 64MB of memory or you'd lose cache. Chipset limitation... 430TX or maybe older, can't recall now.
StevoLincolnite - Wednesday, December 15, 2021 - link
Having 6Gb of Ram doesn't mean it is running in single channel.It means it runs in Asynchronous dual channel mode where the first 4GB is interleaved, but the last 2GB is not.
OFelix - Wednesday, December 15, 2021 - link
"powers of two have been thrown out the window for a while"I'm sorry? What? Do you pass your work through Google Translate 3 times before publishing it?
Dolda2000 - Wednesday, December 15, 2021 - link
Why? I see nothing wrong with that sentence.inighthawki - Wednesday, December 15, 2021 - link
Agreed with Dolda2000 - what is wrong with that sentence? Do you not know basic English idioms?dwillmore - Wednesday, December 15, 2021 - link
This is interesting. While it's true that cell phones and other mobile devices have been using some non-power-of-two memory devices for a while 24GiB devices to stay under the 4GiB limit of some 32bit SoCs, such sizes have never commonly been used on PCs. I'm very concerned with bugs in firmware and in chipsets supporting these sizes well there.Unlike embedded systems where this type of memory has been used, PCs aren't fixed hardware with hardcoded memory sizes (and speed, timings, etc.). Given how many bugs we find in higher level PC software when past assumptions are challenged, I truely fear the many likely latent bugs we're going to find when NPoT memory sizes come to the PC.
I'd sort of feel safer if it came to the server side first, because it's likely to be actually tested there.
Mr Perfect - Wednesday, December 15, 2021 - link
I suspect that as long as the hardware and OS can handle it, the software will be fine. VMs can be assigned non power-of-two ram amounts today. It's something I've made use of myself. When you've got an untold number of VMs all competing for a host's resources, they each get what RAM they need, not what makes their RAM pool a power of two.dwillmore - Wednesday, December 15, 2021 - link
I don't worry about the OS once it gets running. With three bus processors like the older Intel processors (Westmere, etc.), I have no concern with the OS dealing with arbitrary sizes, but the firmware that has to program all the address decoders for the chip select lines, etc. Those I worry about.evanh - Wednesday, December 15, 2021 - link
Most definitely is not a power of two.I'd guess the principle is holes in the physical address space are allowed. And it's up to the MMU to patch the address space back together.
Wrs - Wednesday, December 15, 2021 - link
There's already a hole in the address space whenever you're not using the maximum memory capacity of your controller (usually on the CPU).edzieba - Wednesday, December 15, 2021 - link
Mixing Gibibytes and Gigabits in the same sentence? A bold move!nandnandnand - Thursday, December 16, 2021 - link
AnandTech readers aren't ready for gibibits yet.Dolda2000 - Wednesday, December 15, 2021 - link
Any word on what part of the address is "absorbing" the non-power-of-two? I'd have to guess it's an odd number of bank groups, but who knows what kinds of shenanigans they may be up to.dwillmore - Wednesday, December 15, 2021 - link
No amount of integer multiplication is going to get rid of a factor of three in a value.