By: Jukka Larja (roskakori2006.delete@this.gmail.com), January 1, 2021 9:53 pm
Room: Moderated Discussions
Björn Ragnar Björnsson (bjorn.ragnar.delete@this.gmail.com) on January 1, 2021 8:41 pm wrote:
> Jukka Larja (roskakori2006.delete@this.gmail.com) on January 1, 2021 10:43 am wrote:
> > Gabriele Svelto (gabriele.svelto.delete@this.gmail.com) on January 1, 2021 7:10 am wrote:
> > > me (me.delete@this.me.com) on December 31, 2020 4:56 pm wrote:
> > > > > AMD has their actual server CPU line too, and you do pay more for that privilege, but at least
> > > > > AMD doesn't try to screw you over and limit their non-server parts. So you do get ECC for Threadripper
> > > > > (and plain Ryzen) too, even if it's not necessarily "officially verified".
> > > > >
> > > >
> > > > You would think that for people who want/need ECC, they
> > > > are going to want CPUs that are officially verified.
> > >
> > > What does "officially" mean in this context? All non-APU
> > > Ryzen CPUs support ECC if the motherboards have the
> > > necessary traces and UEFI support. Motherboard vendors advertise this support quite clearly in the specs.
> >
> > Trying to google about how well the unofficial support works, I get lot of hits about people saying that
> > yes, it works, without any proof. I don't see people with a test DIMMs known to produce single bit errors
> > making sure the unofficial support works, or making sure it works in every CPU or at least gives some easy
> > to see error somewhere if it doesn't (I'm sure someone somewhere has tested something, but it gets lost
> > in the noise. Anecdotes are only useful if there's enough of them to be statistically significant).
> >
> > I really like what AMD is doing with CPUs, but unofficial ECC support just
> > annoys me. It's supposed to give me peace of mind and eliminate one source
> > of random problems. "Unofficial" really doesn't work great with that goal.
> >
> > -JLarja
>
> Right now on the AMD side, as previously, we have ECC support for DRAM. ECC capable
> DRAM on the desktop had become nearly extinct as the wast majority of desktop CPUs
> were Intel and didn't support ECC at all, in any shape manner or form.
>
> Now that the "hot" (not thermally speaking) CPUs are from AMD and they do in fact support
> ECC, I have the distinct feeling from my wanderings on the Web that desktop ECC offerings
> are on the rise although I suspect that Crucial/Micron have at the same time been cutting
> back on their unbuffered ECC selection. Strange times indeed.
I didn't have any trouble buying ECC DIMMs for a Linux router / 10G switch I build, but as I only needed about 8 GBs and ended up getting 32 (4 * 8 GB), it didn't really matter too much that the price was nearly the same as 64 GBs (4 * 16 GB) of non-ECC. I think best price/GB would have been with larger 16 GB DIMMs.
> I freely admit that I have not made a rigorous study of this market area recently. My
> excuse is that I'm tired and weary from years and decades of fruitless searching.
>
> I'm fervently hoping that we can see a resurgence in ECC offerings, where, as in the past you
> would be able to buy Parity/ECC memory at every market point at a 5-20% price premium.
>
> Please correct me if I'm wrong.
I just checked one retailer with 175 AM4 motherboards listed. Only 5 were advertising ECC support (actually, only 5 had "ECC" written in description. I presume they all had it because they support ECC DIMMs). None of the 14 TRX40 motherboards had ECC support listed. I have no idea if support is actually that bad or if it's just something manufacturers aren't interested in advertising. Either way, if I was really willing to pay for ECC (both in money and in having to get less suitable motherboard and doing more work), I would look for something with official support.
-JLarja
> Jukka Larja (roskakori2006.delete@this.gmail.com) on January 1, 2021 10:43 am wrote:
> > Gabriele Svelto (gabriele.svelto.delete@this.gmail.com) on January 1, 2021 7:10 am wrote:
> > > me (me.delete@this.me.com) on December 31, 2020 4:56 pm wrote:
> > > > > AMD has their actual server CPU line too, and you do pay more for that privilege, but at least
> > > > > AMD doesn't try to screw you over and limit their non-server parts. So you do get ECC for Threadripper
> > > > > (and plain Ryzen) too, even if it's not necessarily "officially verified".
> > > > >
> > > >
> > > > You would think that for people who want/need ECC, they
> > > > are going to want CPUs that are officially verified.
> > >
> > > What does "officially" mean in this context? All non-APU
> > > Ryzen CPUs support ECC if the motherboards have the
> > > necessary traces and UEFI support. Motherboard vendors advertise this support quite clearly in the specs.
> >
> > Trying to google about how well the unofficial support works, I get lot of hits about people saying that
> > yes, it works, without any proof. I don't see people with a test DIMMs known to produce single bit errors
> > making sure the unofficial support works, or making sure it works in every CPU or at least gives some easy
> > to see error somewhere if it doesn't (I'm sure someone somewhere has tested something, but it gets lost
> > in the noise. Anecdotes are only useful if there's enough of them to be statistically significant).
> >
> > I really like what AMD is doing with CPUs, but unofficial ECC support just
> > annoys me. It's supposed to give me peace of mind and eliminate one source
> > of random problems. "Unofficial" really doesn't work great with that goal.
> >
> > -JLarja
>
> Right now on the AMD side, as previously, we have ECC support for DRAM. ECC capable
> DRAM on the desktop had become nearly extinct as the wast majority of desktop CPUs
> were Intel and didn't support ECC at all, in any shape manner or form.
>
> Now that the "hot" (not thermally speaking) CPUs are from AMD and they do in fact support
> ECC, I have the distinct feeling from my wanderings on the Web that desktop ECC offerings
> are on the rise although I suspect that Crucial/Micron have at the same time been cutting
> back on their unbuffered ECC selection. Strange times indeed.
I didn't have any trouble buying ECC DIMMs for a Linux router / 10G switch I build, but as I only needed about 8 GBs and ended up getting 32 (4 * 8 GB), it didn't really matter too much that the price was nearly the same as 64 GBs (4 * 16 GB) of non-ECC. I think best price/GB would have been with larger 16 GB DIMMs.
> I freely admit that I have not made a rigorous study of this market area recently. My
> excuse is that I'm tired and weary from years and decades of fruitless searching.
>
> I'm fervently hoping that we can see a resurgence in ECC offerings, where, as in the past you
> would be able to buy Parity/ECC memory at every market point at a 5-20% price premium.
>
> Please correct me if I'm wrong.
I just checked one retailer with 175 AM4 motherboards listed. Only 5 were advertising ECC support (actually, only 5 had "ECC" written in description. I presume they all had it because they support ECC DIMMs). None of the 14 TRX40 motherboards had ECC support listed. I have no idea if support is actually that bad or if it's just something manufacturers aren't interested in advertising. Either way, if I was really willing to pay for ECC (both in money and in having to get less suitable motherboard and doing more work), I would look for something with official support.
-JLarja