This is supposed to be the final verdict: https://www.theverge.com/2024/10/4/24262287/intel-13th-14th-gen-crash-raptor-lake-root-cause-fix
Intel is still not publicly providing the batch numbers or serial number ranges of Raptor Lake chips that suffered from oxidization during manufacturing — it claims it fixed that issue long ago — and it does not have an update on a tool to let you easily test your chip to see if it’s aged prematurely. Intel is also still not providing an estimate of how many chips are likely to be irreversibly impacted, or how lenient it will be with warranty support.
I personally do not trust anything Intel has said lately. Until that changes, I will not buy any Intel products.Quoting the article:
. . . yeah, that's awesome.
Anyone looking at Intel consumer products (particularly DiY desktop) should be thinking twice before giving them any money. NV and their awesome melting 12VHPWR connectors were bad enough, but this debacle has been much worse.I personally do not trust anything Intel has said lately. Until that changes, I will not buy any Intel products.
Disagree, melting connectors are a safety issue. Your house could burn down.NV and their awesome melting 12VHPWR connectors were bad enough, but this debacle has been much worse.
What is this whitewashing bs? Reddit, forums, and youtube are filled with raptor owners horrible experiences with Intel RMA service, including rejections. That is, once they knew the CPU was broken and the source of their problems. Before that it was months and months of crickets chirping while Intel let game devs and gpu vendors take the heat for games crashing. Throwing the board makers under the bus. Selling CPUs in 2024 they knew had via oxidation in 2022.What is gonna happen when an Intel CPU quits working? They (eventually) give you a newer better one?
Yes this is also bad, but it was contained to mostly RTX 4090 users. I have 4070 Super with this connecter its fine and I made sure it clicked. This was only an issue on the RTX 4090 and some RTX 4080 cards. The 4070 Ti and below cards don't pull enough power to melt the cable otherwise this would been an even bigger issue as most Ada owners with the 12VHPWR connectors are 4070 Super owners.Disagree, melting connectors are a safety issue. Your house could burn down.
Disagree, melting connectors are a safety issue. Your house could burn down.
What is gonna happen when an Intel CPU quits working? They (eventually) give you a newer better one?
Kinda true but the supposedly newer chip could've the same issue. And all the lost time troubleshooting which would translate into real money if you are a busy professional. The Intel debacle is arguably worse due to the sheer number of users impacted, some of whom even ran their CPUs at pedestrian speeds to avoid crashes. We have at least two users on this very forum (Hulk and Kocicak) who actually "discovered" the issue on their own but no one paid attention because who could really fathom that Intel's incompetence could reach such new heights?What is gonna happen when an Intel CPU quits working? They (eventually) give you a newer better one?
I don’t get what you’re getting at? What’s the cynical motive?
The fix reduces the performance in some benchmarks so ARL turns out looking a lot better than if RPL worked flawlessly and hit the intended speeds. Some impacted users may also say F it and just get Arrow Lake because in their mind, it's just a hiccup and they will forgive Intel if ARL works fine. And they can recoup some of their lost investment by selling their one or two year old PC to some sucker who will next try to be forgiving to Intel's follies by being careful not to push the RPL CPU too hard.I don’t get what you’re getting at? What’s the cynical motive?
That’s not the case though and even if it were it’s not like you couldn’t just test the pre patched microcode.The fix reduces the performance in some benchmarks so ARL turns out looking a lot better than if RPL worked flawlessly and hit the intended speeds.
Let's see how many reviewers actually do exactly thatThat’s not the case though and even if it were it’s not like you couldn’t just test the pre patched microcode.
Do you have any evidence that microcode 0x12B is a performance regression compared to 0x129?Let's see how many reviewers actually do exactly that
Here's a nice example of the final fix wrecking potential perf by disallowing higher speeds: https://www.techpowerup.com/forums/...update-gimped-wrecked-my-rl-overclock.327286/Do you have any evidence that microcode 0x12B is a performance regression compared to 0x129?
When Arrow Lake launches, they can bench it against "nu Raptor Lake" and show bigger gains than against launch Raptor Lake. You know, the performance it was meant to exhibit for its lifespan?I don’t get what you’re getting at? What’s the cynical motive?
Just bringing this up again after it’s left the news cycle is bad. The tech press moved on and the new microcode has more or less quieted the loudest criticisms. From a Machiavellian standpoint, your best move is to say absolutely nothing at this point.
Edit: Adding onto this, they just released Lunar Lake and Granite Rapids and both are doing really well in their segments. Why would they want to turn attention to the PR disaster that is Raptor Lake?
I don't see how this works with AMD CPUs being the main comparison point. AMD also got the perf boost recently, which will hurt RPL even more.When Arrow Lake launches, they can bench it against "nu Raptor Lake" and show bigger gains than against launch Raptor Lake. You know, the performance it was meant to exhibit for its lifespan?
But most Intel users, similar to Nvidia and Apple users, just avoid doing that. Just look at the users who suffered for MONTHS with crashes, blaming software and other hardware excluding the CPU and the mobo makers were thrown under the bus by Intel who should've dictated to them the acceptable parameters for running their CPUs error free. The fact that even stock operation of the CPUs was affected, is just unforgivable. I wouldn't be happy with this situation if I were a Raptor Lake owner unless Intel promised to replace my CPU and mobo with a brand new Arrow Lake and compatible mobo. May seem like unreasonable and excessive expectation but Intel should consider the alternative: I could turn to AMD and never return.People seeing the 14900K presumably struggling against Zen 4 would start asking themselves some uncomfortable questions.
Well, let's review sites constantly use blue lines for Intel and red lines for AMD, then and that those who only buy blue have these red-filtering glasses:I don't see how this works with AMD CPUs being the main comparison point. AMD also got the perf boost recently, which will hurt RPL even more.
People seeing the 14900K presumably struggling against Zen 4 would start asking themselves some uncomfortable questions.
So we just forget the first few months for AMDs silliness as well when Zen5 launched.But most Intel users, similar to Nvidia and Apple users, just avoid doing that. Just look at the users who suffered for MONTHS with crashes, blaming software and other hardware excluding the CPU and the mobo makers were thrown under the bus by Intel who should've dictated to them the acceptable parameters for running their CPUs error free. The fact that even stock operation of the CPUs was affected, is just unforgivable. I wouldn't be happy with this situation if I were a Raptor Lake owner unless Intel promised to replace my CPU and mobo with a brand new Arrow Lake and compatible mobo. May seem like unreasonable and excessive expectation but Intel should consider the alternative: I could turn to AMD and never return.
I go back to the ti-99 era. Was also an Amiga guy until they stopped porting games to the platform. The bleeding edge has always been a thing. You don't early adopt unless you like chancing the pain. The old rule of thumb was wait at least 6 months. Complaining about new stuff being janky is silly IMO; it's the same as it ever was.So we just forget the first few months for AMDs silliness as well when Zen5 launched.
Yes, it’s doesn’t pale in comparison to Intels stupidity but I’m disliking this duopoly more and more as time goes on. Maybe the US government should open up x86 so more companies can design x86 CPUs instead of funding Intel. It’s sucks that we are locked in to the most insane roller coaster companies for CPUs.
I go back to the ti-99 era. Was also an Amiga guy until they stopped porting games to the platform. The bleeding edge has always been a thing. You don't early adopt unless you like chancing the pain. The old rule of thumb was wait at least 6 months. Complaining about new stuff being janky is silly IMO; it's the same as it ever was.
Raptor Lake however, is completely different and much worse. In hindsight, it is a symptom of a company that's is unhealthy/rotting away from the inside. Had it been handled properly the stink would have washed off by now. As is, it smells like gangrene, hence the amputations taking place.
That's all on Pat. How dare he let that happen under his watch???According to Wendell, there were veteran engineers at Intel expressing concern and warnings about what was going on, but they were ignored by management until everything started to blow up publicly.
That's all on Pat. How dare he let that happen under his watch???