AMD's Ryzen CPUs (Ryzen/TR/Epyc) & Vega/Polaris/Navi GPUs

Yes the same T series and P series Slim series. Although pricing is different in India and starts at 1300$ base specs and Intel's is close to 1700-1800$ base specs.

 

Not all see it this way. Maybe Puget Systems is all too small to provide correct info. And from what I saw the 12th gen is most likly too new to provide any reliable statistics for the comparison. But the differences between Red and Blue ain't big.

Puget data suggests Intel CPUs generally were more reliable than AMD's in 2019-21 neowin · Jan 13, 2022

Data from the last two years on AMD and Intel Ryzen, Core, Threadripper, and Xeon CPUs shows that the processors generally from the Team Blue were more reliable than its rival Red team ones.

 

Interesting that they would arrive to that conclusion when their data shows that the 11th gen was >2x more likely to fail in the shop and in the field than the 5000 series from AMD...

Perhaps they didn't know how to analyze the data?

 

I think they took the analyze for more than one SKUs in their lineups. 3 from Red and 5 from the Blue team. They build different setups from what I can see from the article. And they offer analyze from two different fields.

Edit.
Btw. 11th gen was a disaster with the backported technologie. A disgusting end for Socket 1200. I have already mentioned this in the desktop thread. Not sure if we can put the 11th gen BGA on 10nm into this.

 

I know they analyzed more than one SKU, however it's funny they would make such a broad statement when there is clearly more to the story. If the 5k series from AMD and the 11th gen were the biggest sellers in their analyzed products (pretty safe to say), I would say that AMD is in a much better position there, and likely makes up for that.

I haven't had a failed CPU in a number of years (and that was my fault when doing some extreme overclocking), but my sample size is small, only a couple hundred.

 

Not sure how many builds they have done with 11th gen. And the failure rates of various hardware brands and models is done over the last three years. 11th gen was released March 30, 2021 (9 months ago).

While this is normally an internal process, sometimes we like to give our customers and readers a peek into aspects of our company that aren't usually in the public spotlight. In this article, we will look at both shop (caught in our production process) and field (after the system has shipped to the customer) failure rates of various hardware brands and models over the last three years. At the end, we will crown the most reliable component used in our workstations.

Their comments... AMD CPUs in general had higher failure rates than Intel, but we did see an oddly high rate of failures with Intel's consumer-oriented 11th Gen processors... which seems odd, especially next to the very low rates shown by the preceeding 10th Gen.

Source and image: Puget

 

Their very last sentence indicates that they are biased, lol. It sounds like they are just sweeping that under the rug.

It's all good though, I honestly don't think that CPU failure rates is anything people should worry about since it happens so rarely. As I swap between my AMD personal laptop (thinkbook) and my Intel work laptop (HP), I appreciate no difference outside of the fan on the stupid HP is always on and super annoying.

 

Funny enough, I didn't expect anything less. ( Total dismissal of the facts).

This is why there is no point offering them here.

Thanks @Papusan, you went above and beyond, with calming patience. But even that isn't enough for the biased and the fanbois.

 

Please continue to insult me and call me a fan boy when I clearly stated my case. Please feel free to come up with your own opinion when looking at the data that clearly shows >2x failure rates for 11th gen over the 5k series. Also, just curious if you recalled that insulting anyone on these forums is against forum rules, that includes moderators...

*on edit* also I must be the worst fan boy given my signature... What can I say, I like to fiddle.

 

It's surprising to see heated AMD vs Intel debates. After all, they are both supliers of PC CPUs. It is clearly in the PC users' interest to have these two interlocked in eternal competitive struggle and a state of equilibrium. Consequently, it is rational for these consumers to at least motally support the underdog if any. If in doubt regarding who the weaker side might be, I guess we could look at the current say PC laptop CPU market share data. I guess it's still AMD catching up, if so then go AMD!

 

Now, a couple points against AMD during this time period. When Zen 3 rolled out (or Zen 2, I forgot which), puget initially found a number of amd cpus DOA. It was right after the announcement. The first batches seemed to have this issue. They also had MB mfrs pushing insanely high voltage into AMD cpus also at that time. This could severely effect perception over the life span.

Further, AMD during this time period has severe USB issues.

Just noting quality of life and growing pains which could effect perception.

Also, one needs to not just look at the number dead, but the number deployed to have a failure rate per number of CPUs to help make it a fair comparison.

Note- I have not reviewed Puget's page yet. I'm on a break. Instead, giving known issues that may have influenced the numbers to give context.

 

Their analysis only includes the following:

• AMD Ryzen 5000 Series
• AMD Threadripper 3000 Series
• AMD Threadripper Pro 3000 Series
• Intel Core 10th Gen
• Intel Core 11th Gen
• Intel Core X 10000 Series
• Intel Xeon W 2200
• Intel Xeon Scalable 2nd Gen

They do calculate failure rate in terms of "shop failure rate" and "field failure rate." It clearly indicates that the 11th generation Intel had twice the rate of failures as the AMD Ryzen 5000 series. It's a pretty simple read when you do get around to it. I do understand the initial issues that AMD had, I seem to be having similar new tech pains with my LGA1700 systems...

 

I just took a peak and the failure rate, if you exclude the 11th Gen, would be more heavily on AMD.

But this is over a 3 year span. And having a 6.6% failure rate on any line of CPU is huge! AMD's was claimed that high at first with the Zen 3 launch on failures with system integrators, but I think debauer did a video discussing he was seeing less than that with caseking.

But Intel has cover for this year due to supply shortages.

So, over the life span of the platform, we might see that percentage drop, but it seems they made their conclusion by excluding that line of chips. Which, if excluded, I would have come to the same conclusion.

Edit: BTW, 5% is a 1 in every 20 chips failing. Intel's chips for the 11th gen was failing at a higher rate than that!

 

Yes if you exclude the direct competitor to the AMD 5000 series then AMD did have a higher failure rate. What would happen if you excluded that generation of AMD cpus? We don't know because we don't have the data available. You can't just exclude findings when they don't fit your narrative.

*on edit* - that was my whole issue with the article. If they take out the glaring failures then yes they can make that assessment, but is that really going by the evidence?

 

I don't think it was Puget Systems that point it out first or ever.... https://wccftech.com/amd-ryzen-5000-zen-3-desktop-cpus-x570-motherboards-high-failure-rates/

Custom PC Builder PowerGPU Claims Ryzen 5 5000 Zen 3 CPUs Experiencing High Failure Rates

 

Exactly my point, sure they had high failure rates, but half that of Intel's 11th generation...

 

It was to that article that der8aur responded. He was seeing elevated rates, but much lower than that. Also, that is a small sample size. Shortly after that, people discovered motherboard manufacturers running nearly 1.5V into the CPU at idle for stock bios. It was quickly resolved. In that same 10 month time period, intel's 11th gen has had over 6% failure rate, while AMD saw much lower.

So pointing out that article, addressed shortly after, without the full context, is a bit asinine at this point.

Edit: but that incident is what I was talking about. And even with that, puget saw a lower percentage failure rate than Intel.

 

Remember the analyze is for over 3 years span. 11th gen have been out in 9 months.

This is a bigger problem for AMD. Increased prices and almost none refreshed chips before end of 2022. Even worse than the XT line chips (offered double up for each SKU). Aka they struggle now to have a real upgrade path.

TSMC’s 3D Tech Supply & Manufacturing Woes May Cause Limited AMD Ryzen 7 5800X3D Availability, Could Also Explain No 3D Variants of 5900X & 5950X

 

AMD 5000 series has been out for less than 14 months, what's your point?

 

You do realize at the end of the year, zen 4 drops, right?
https://www.pcgamer.com/amd-zen-4-ryzen-7000-first-glimpse-ces-2022/

So, to be fair, they will still sell out their silicon all year because shortages will continue (meaning no meaningful loss of sales due to competition from intel's 12th gen), and they don't stuff the channels with products they can't clear once AM5 drops.

Edit: they also are using tsmc supply as a shield, saying it isn't our fault we are primarily putting the zen3d chips into higher margin server products, it's tsmc's supply.

It's a BS line from AMD to take the heat off of them for them doing a greedy move to make more money.

 

What I mean... Offer only Ryzen 7 5800X3D as a final nail for AM4 is equal disgusting as Rocket lake was for socket 1200. And read or see what F. Azor said on this topic. Cost, cost and cost. Why offer a upgrade if the new tech is too expensive.

And yep. The screwed one can get the upgrade now. Why bother with those that hoped for a happy ending of AM4 on the newest boards.
ASRock First to Support AMD Ryzen 5000 CPUs on X370 Chipsets

 

Yes and no. The zen3d they are opening up to x370 boards, meaning that they can sell it even if they don't buy new motherboards, likely as a way to clear inventory to prevent it sorting on shelves because zen 3 owners won't upgrade.

Meanwhile, Intel did an odd socket stepping stone that made little sense. They didn't plan well. But, to be fair, aside from MB fires, their 12th gen seems to win some tasks, with an asterisk for others.

Either way, I'm mostly done with OCing, so I don't care, it's best value and with product on shelves when I buy. Other than that, I couldn't care less who I buy from.

I'm sure they just wanted more money from the server segment (see edit on my comment before this one).

 

Yep, they clear out inventory. Aka no love for those that hoped for a final new chips for their sub 1 year old EOL platforms. All they get is an lower clocked gaming chips that perform better for games. Almost as a side-grade outside gaming. Oh'well. Intel have been there, now also AMD

 

Definitely agree. This is why when they shafted x399 owners, I said I wouldn't buy AMD until Zen4 and I meant it. That was the next reasonable upgrade path for me.

When my Skylake machine died, there was no amd mainstream chips to be found. So I bought intel's 10700kf. Still need a gpu upgrade.

Now, my new hobby is doing simulations of speaker enclosures and designing active speakers. I have a tower design using scans scanspeak subs, then seas woofers and tweeters. I've designed bookshelf speakers with sb acoustic tweeters and scanspeak woofers. I even have an early design of 2 8" Dayton audio subs paired with sb acoustic woofers and tweeters.

I'm buying a carton of ice icepower amplifiers (400A2 amps) if anyone is interested (primarily to US people because import vat would likely kill the value of picking it up from me). Then, I'm using Arylic up2stream boards feeding into a digital signal processor (ADSP-21489 with burr brown ADC and DAC chips).

But that is my new hobby. I just use simulations to optimize the designs nowadays. Part of those simulations is highly serial, meaning 1-2 cores loaded.