AMD's Ryzen 7 7800X3D Selling Nearly Twice as Fast as 5800X3D in Some Regions

[mechtech]

ya in regions with monies no doubt at all, especially with ddr5 coming down in price.

 

[Space Lynx]

hope they fix their problems very fast, so not many customers get disappointed

also, a lot of customers aren't like us on this website, they won't know anything about BIOS, what it even is, that they need to update their BIOS, etc.

hopefully no chips die.

 

[jdgesmlls]

This has happened with setups running with default settings aswell.

Yes, Gamers Nexus proved that this "can" happen with default settings but the question is under what conditions/how often? The discussion is about how widespread the problem is using the total number of occurrences vs. the number of chips sold. Those percentages are never discussed. The sensational images of burning CPUs and motherboards in the video are shared, and all discussion gets lost. Gamer Nexus then generalized the entire AM5 platform as "a mess" and their viewers parrot those opinions.

I have a B650E Strix, purchased with a 7900x and swapped to a 7800x3D... only updating the BIOS once to support the X3D chip (prior to the slew of BIOS updates when the voltage issue was discovered). I have no issues with HW. The 7900x chip is pristine, the 7800x3D is pristine, the motherboard is pristine... So what is special about my setup? Why didn't I encounter the issue? This is why the criteria and likelihood of occurrence is important.

 

[evernessince]

That has to be purchases made prior to the whole meltdown debacle, cause the longevity of these cpu's are seriously in doubt following that... i for one am gonna replace it before warranty runs out.

Is it? Record sales and we have less than a handful of examples of CPUs actually experiencing the issue. Ryzen 7000 series in general has a lower RMA rate than the 5000 series.

Bound to happen.

Fastest gaming CPU in the world, latest platform, latest technologies. Seems that the premium in price is worth it to many people after all.

The 5800X3D ticked all those boxes as well. It may be the low power consumption that makes the 7800X3D so appealing, a lot of people seem to discount it as a significant factor but for many it's a huge factor.

Yes, Gamers Nexus proved that this "can" happen with default settings but the question is under what conditions/how often? The discussion is about how widespread the problem is using the total number of occurrences vs. the number of chips sold. Those percentages are never discussed. The sensational images of burning CPUs and motherboards in the video are shared, and all discussion gets lost. Gamer Nexus then generalized the entire AM5 platform as "a mess" and their viewers parrot those opinions.

I have a B650E Strix, purchased with a 7900x and swapped to a 7800x3D... only updating the BIOS once to support the X3D chip (prior to the slew of BIOS updates when the voltage issue was discovered). I have no issues with HW. The 7900x chip is pristine, the 7800x3D is pristine, the motherboard is pristine... So what is special about my setup? Why didn't I encounter the issue? This is why the criteria and likelihood of occurrence is important.

GN already answered the question about how widespread the issue is, they stated that the issue is exceedingly difficult to reproduce and only happens if you "roll a 1 on the motherboard, CPU, and RAM". In otherwords you have to get thrice unlucky to an extreme degree. You have to get a CPU on the very low end of the silicon quality spectrum and a motherboard willing to massively overvolt that CPU.

also, a lot of customers aren't like us on this website, they won't know anything about BIOS, what it even is, that they need to update their BIOS, etc.

hopefully no chips die.

Maybe ASUS does something different but every other mobo vendor I'm aware of runs the CPUs at stock voltages, aka 1.01v for stock SOC. I would be for a warranty extension for any potentially impacted CPUs though, just in case.

 

[Redwoodz]

That has to be purchases made prior to the whole meltdown debacle, cause the longevity of these cpu's are seriously in doubt following that... i for one am gonna replace it before warranty runs out.

Replace it how? Send it back for no reason? Funny how there are no 58003XD users complaining. They had the voltage locked at 1.3v from the start. Every Ryzen cpu has had a theoretical 1.3v limit. If your motherboard doesn't exceed AMD's limit then you will have zero issues. How hard is that to understand? The fact is some mobo man. went beyond to boost "stock" memory speeds, so guys like you can say only Asus has the best memory performance...yeah right.

 

[jdgesmlls]

GN already answered the question about how widespread the issue is, they stated that the issue is exceedingly difficult to reproduce and only happens if you "roll a 1 on the motherboard, CPU, and RAM". In otherwords you have to get thrice unlucky to an extreme degree. You have to get a CPU on the very low end of the silicon quality spectrum and a motherboard willing to massively overvolt that CPU.

So not only is the problem rare according to Gamers Nexus, there is no way a typical end user can easily rate their silicon quality to determine the likelihood of encountering the issue. That doesn't "answer" anything.

 

[Easo]

Good for AMD, not sure what else to add.

 

[Daven]

Thank you everyone else for your posts about this rare and difficult to produce issue. I was originally under the impression it wasn’t something widespread and it looks like that is the case.

I guess Gamernexus got a lot of clicks including one from me for that video.

 

[Space Lynx]

Is it? Record sales and we have less than a handful of examples of CPUs actually experiencing the issue. Ryzen 7000 series in general has a lower RMA rate than the 5000 series.



The 5800X3D ticked all those boxes as well. It may be the low power consumption that makes the 7800X3D so appealing, a lot of people seem to discount it as a significant factor but for many it's a huge factor.



GN already answered the question about how widespread the issue is, they stated that the issue is exceedingly difficult to reproduce and only happens if you "roll a 1 on the motherboard, CPU, and RAM". In otherwords you have to get thrice unlucky to an extreme degree. You have to get a CPU on the very low end of the silicon quality spectrum and a motherboard willing to massively overvolt that CPU.



Maybe ASUS does something different but every other mobo vendor I'm aware of runs the CPUs at stock voltages, aka 1.01v for stock SOC. I would be for a warranty extension for any potentially impacted CPUs though, just in case.

enabling expo triggers the SOC to go up though, that was the issue all along I thought?

the common consumer knows about xmp and expo imo, so its still an issue

 

[evernessince]

So not only is the problem rare according to Gamers Nexus, there is no way a typical end user can easily rate their silicon quality to determine the likelihood of encountering the issue. That doesn't "answer" anything.

Them not being able to check their silicon quality is irrelevant. If they are aware of the potential for the issue, a simple BIOS update is all that's needed. And again, silicon quality is only a single factor. You need to get unlucky with the CPU, Motherboard, and RAM in addition to not updating BIOS. GN said the issue was exceedingly difficult to reproduce for a reason.

 

[eidairaman1]

Ofc its the most popular due to price

 

[evernessince]

enabling expo triggers the SOC to go up though, that was the issue all along I thought?

the common consumer knows about xmp and expo imo, so its still an issue

EXPO doesn't set the SOC voltage, the motherboard does. Yes this issue happens in rare cases with EXPO enabled but that's down to some motherboards setting far too high a voltage, not from anything EXPO does. This is mostly limited to ASUS boards wherein they set a voltage of 1.4 to 1.45V, which stock for comparison is 1.01v. My X670 Taichi for comparison sets SOC voltage to 1.24v with EXPO enabled in competitive subtimings mode. If a customer is aware of EXPO, they should know about updating their BIOS which would render the whole thing a non-issue in any case as newer BIOSes that address the issue have been out.

 

[Space Lynx]

If a customer is aware of EXPO, they should know about updating their BIOS which would render the whole thing a non-issue in any case as newer BIOSes that address the issue have been out.

I disagree, a lot of general consumers know about xmp/expo because a friend will just shout randomly, oh hey dont forget to enable xmp so your ram is running at advertised speeds, but bios are rarely ever talked about among common customers i come across anyway, or bios is considered taboo and to not be messed with

 

[MicroUnC]

Yes, Gamers Nexus proved that this "can" happen with default settings but the question is under what conditions/how often? The discussion is about how widespread the problem is using the total number of occurrences vs. the number of chips sold. Those percentages are never discussed. The sensational images of burning CPUs and motherboards in the video are shared, and all discussion gets lost. Gamer Nexus then generalized the entire AM5 platform as "a mess" and their viewers parrot those opinions.

I have a B650E Strix, purchased with a 7900x and swapped to a 7800x3D... only updating the BIOS once to support the X3D chip (prior to the slew of BIOS updates when the voltage issue was discovered). I have no issues with HW. The 7900x chip is pristine, the 7800x3D is pristine, the motherboard is pristine... So what is special about my setup? Why didn't I encounter the issue? This is why the criteria and likelihood of occurrence is important.

Do you use default settings or with EXPO? I've just built mine and use pre beta bios. I am kinda scared to use EXPO.

 

[Dragam1337]

I disagree, a lot of general consumers know about xmp/expo because a friend will just shout randomly, oh hey dont forget to enable xmp so your ram is running at advertised speeds, but bios are rarely ever talked about among common customers i come across anyway, or bios is considered taboo and to not be messed with

Agreed, bios is something the average user will never upgrade. I'd like to see some statistics on this, but im gonna guess that everyone with a prebuilt pc (vast majority, sadly) wont be upgrading bios.

 

[doc7000]

That has to be purchases made prior to the whole meltdown debacle, cause the longevity of these cpu's are seriously in doubt following that... i for one am gonna replace it before warranty runs out.

The cause of damaged CPUs is now understood, and motherboard makers are guilty here.

 

[Dragam1337]

Yes, Gamers Nexus proved that this "can" happen with default settings but the question is under what conditions/how often? The discussion is about how widespread the problem is using the total number of occurrences vs. the number of chips sold. Those percentages are never discussed. The sensational images of burning CPUs and motherboards in the video are shared, and all discussion gets lost. Gamer Nexus then generalized the entire AM5 platform as "a mess" and their viewers parrot those opinions.

I have a B650E Strix, purchased with a 7900x and swapped to a 7800x3D... only updating the BIOS once to support the X3D chip (prior to the slew of BIOS updates when the voltage issue was discovered). I have no issues with HW. The 7900x chip is pristine, the 7800x3D is pristine, the motherboard is pristine... So what is special about my setup? Why didn't I encounter the issue? This is why the criteria and likelihood of occurrence is important.

There is more than one b650e strix... is the a, e, or f ?

I have a b650e-f, and i smelled a strong smell of burned electronics coming from my pc the other day, which has been followed by increasing amount of system instability. My guess is that the chip is at least semi fried, but i just haven't had the energy to pull it apart.

And this was while using default settings, no expo.

The cause of damaged CPUs is now understood, and motherboard makers are guilty here.

No, cause if it was just expo voltage that was the issue, then no cpu would have died with default settings, which some have...

 

[evernessince]

I disagree, a lot of general consumers know about xmp/expo because a friend will just shout randomly, oh hey dont forget to enable xmp so your ram is running at advertised speeds, but bios are rarely ever talked about among common customers i come across anyway, or bios is considered taboo and to not be messed with

Agreed, bios is something the average user will never upgrade. I'd like to see some statistics on this, but im gonna guess that everyone with a prebuilt pc (vast majority, sadly) wont be upgrading bios.

No, updating BIOS on AMD motherboards is definitely a lot more common than you think. The number of people updating BIOS just to get newer CPU support is enormous. A lot of people buying into the AM5 platform are doing so explicitly with the intent to update BIOS later to get newer CPU support.

You might have a point if this were an Intel platform but given AM5 will have 3 generations of CPU support the number of people upgrading the first generation of boards is extremely high.

 

[Dragam1337]

No, updating BIOS on AMD motherboards is definitely a lot more common than you think. The number of people updating BIOS just to get newer CPU support is enormous. A lot of people buying into the AM5 platform are doing so explicitly with the intent to update BIOS later to get newer CPU support.

You might have a point if this were an Intel platform but given AM5 will have 3 generations of CPU support the number of people upgrading the first generation of boards is extremely high.

Among system builders*

Most people buy a prebuilt, and do you honestly think that people who buys a prebuilt upgrade the bios ?

 

[Space Lynx]

No, updating BIOS on AMD motherboards is definitely a lot more common than you think. The number of people updating BIOS just to get newer CPU support is enormous. A lot of people buying into the AM5 platform are doing so explicitly with the intent to update BIOS later to get newer CPU support.

You might have a point if this were an Intel platform but given AM5 will have 3 generations of CPU support the number of people upgrading the first generation of boards is extremely high.

Alright, I did kind of forget about the bios updates for newest cpu, that has been shoved in peoples faces, even stickers on boards saying it has the latest bios for this new chip, etc. So ok, I can concede. I think thanks to the AM4 platforms entire life cycle you are correct on this actually. Before then I would disagree with you though.

 

[Vayra86]

Might quadruple facepalm at that spelling. Wow! Typical gen z? I wrote it down, I re read it, it doesn't look right, but whatever!

Ontopic; I'm actually quite surprised at how well these things sell, especially given the expo bug. Its also not the cheapest CPU... Damn.

Among system builders*

Most people buy a prebuilt, and do you honestly think that people who buys a prebuilt upgrade the bios ?

Yeah, I think its absolutely silly what motherboard vendors do wrt CPUs, I mean... OC out of the box - or auto OC voltages - is just a big fat nono in my book, regardless of whatever segment you're making products for. Any tweak they do on a board should be clearly and strongly supported by clear notifications - and not written in Chinglish with info missing.

 

[jdgesmlls]

Do you use default settings or with EXPO? I've just built mine and use pre beta bios. I am kinda scared to use EXPO.

I've used EXPO since day 1 and have never disabled it, nor adjusted any SOC voltages.

There is more than one b650e strix... is the a, e, or f ?

I have a b650e-f, and i smelled a strong smell of burned electronics coming from my pc the other day, which has been followed by increasing amount of system instability. My guess is that the chip is at least semi fried, but i just haven't had the energy to pull it apart.

And this was while using default settings, no expo.



No, cause if it was just expo voltage that was the issue, then no cpu would have died with default settings, which some have...

B650E-F.
Good.... report back on how the RMA process is being handled by AMD/Asus...

 

[trsttte]

interesting ...

oh well i will probably be contributing to the 5800X3D side soon ... these are nearing the sub 300chf/$ mark

Same here, i've been tracking it in geizhals and other than a flash sale a month ago seems pretty stable at around 310€

Maybe if they drop desktop phoenix apus soon prices move a bit, otherwise it's looking like a bottom

 

[evernessince]

Among system builders*

Most people buy a prebuilt, and do you honestly think that people who buys a prebuilt upgrade the bios ?

I like the angle and I'm sure not many people considered OEM systems. That said if OEM systems are indeed the bulk of sales and yet we've not had a single report of the issue cropping up on them then that's an indicator that either the issue doesn't exist at all on those systems or it happens at a greatly reduced rate (and that's considering the issue already happens at a very very low rate to begin with). Remember the issue ultimately stems from too much SOC voltage, so it's going to depend on the motherboard used.

I think what we really do need for customers though is an extension of warranty and guarantee of a quick replacement at the customer's discretion for any CPU installed on a motherboard that set SOC above 1.3V automatically with EXPO enabled and that was sold before the BIOS fix was implemented.

I've used EXPO since day 1 and have never disabled it, nor adjusted any SOC voltages.


B650E-F.
Good.... report back on how the RMA process is being handled by AMD/Asus...

Same. SOC stays at a steady 1.24v. This is with DDR5 6000 CL30.

 

[Dragam1337]

I like the angle and I'm sure not many people considered OEM systems. That said if OEM systems are indeed the bulk of sales and yet we've not had a single report of the issue cropping up on them then that's an indicator that either the issue doesn't exist at all on those systems or it happens at a greatly reduced rate (and that's considering the issue already happens at a very very low rate to begin with). Remember the issue ultimately stems from too much SOC voltage, so it's going to depend on the motherboard used.

I think what we really do need for customers though is an extension of warranty and guarantee of a quick replacement at the customer's discretion for any CPU installed on a motherboard that set SOC above 1.3V automatically with EXPO enabled and that was sold before the BIOS fix was implemented.



Same. SOC stays at a steady 1.24v. This is with DDR5 6000 CL30.

The kind of people that buys prebuilts wouldn't have taken it apart and found this issue, nor would they even frequent these kinda hardware sites - they would have called the shop they bought it from, who would then quietly have swept it under the rug, and just replaced the machine.