GoldenTiger
Fully [H]
- Joined
- Dec 2, 2004
- Messages
- 29,929
It was, if I recall correctly, water cooling, nothing exotic .they had their chip hooked up to some exotic cooling
Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
It was, if I recall correctly, water cooling, nothing exotic .they had their chip hooked up to some exotic cooling
BBC Model B with 6502 @ 2MHz says hello, the secondary 6502 expansion at 3MHz gives a fist pump.there are processors from the 70's that still work fine. these intel chips aren't lasting 3 months. that's how we can tell
It was, if I recall correctly, water cooling, nothing exotic .
Hope it doesn't put them too far under to keep competing with AMD
It was an icewater chiller, not just water cooling.It was, if I recall correctly, water cooling, nothing exotic .
A bit simplified, but this we can agree upon. Temps and current needs especially to be in check to avoid problems from electron migration. Outside edge cases, it should not be a problem during the lifetime of the CPU if run within acceptable limits. But my point still stands. You cannot state that there has been no degradation without specialized tools. That a system is stable, does not mean there is no degradation. System stability is not a proof of no degradation and the only thing you can state is that you have not observed any point of failures due to degradation (or you only played Solitare and not exposed the CPU to any scenario where you can observe this)....
Keep vcore and temps within acceptable limits and electromigration simply isn't a problem.
Yes. People are observing failures occurring more and more frequent. Some have changed CPU and problems went away, others have reduced clocks and/or used intel failsafe voltage to stabilize their systems. It seems to occur over time and start out stable. This smells like bad overclock @ stock from MB manufacturers. You can easier tell that there has been a degradation without specialized tools, then to say that there is no degradation which was my point to Mazzspeed.there are processors from the 70's that still work fine. these intel chips aren't lasting 3 months. that's how we can tell
Yes exactly. It was definitely a low point for Intel. They got caught out for running absolutely exotic cooling with huge power requirements. So funnyIt was an icewater chiller, not just water cooling.
Well if that's the case, than you can't claim electromigration can be a problem during the lifespan of a CPU without specialized tools. So checkmate, my point still stands.A bit simplified, but this we can agree upon. Temps and current needs especially to be in check to avoid problems from electron migration. Outside edge cases, it should not be a problem during the lifetime of the CPU if run within acceptable limits. But my point still stands. You cannot state that there has been no degradation without specialized tools.
Moving the goalpost much? My claim is that you cannot claim that you have no degradation or no extra degradation due to overclock without any specialized tools to observe that. And I dont claim electron migration can be a problem during the lifespan of a CPU without specialized tools. That is moving the goalpost.Well if that's the case, than you can't claim electromigration can be a problem during the lifespan of a CPU without specialized tools. So checkmate, my point still stands.
Except I don't need specialized tools to observe the possibility of operational degradation regarding the CPU's in question, which was my point from the beginning. No goal posts were shifted whatsoever.Moving the goalpost much? My claim is that you cannot claim that you have no degradation or no extra degradation due to overclock without any specialized tools to observe that. And I dont claim electron migration can be a problem during the lifespan of a CPU without specialized tools. That is moving the goalpost.
You claimed no observable degradation. You need specialized tools to do such observation...
Saying your system is stable, does not prove no degradation.
Yes you do if you are to make claim your CPU have no degradation. The others observe a point of failure and it points to a degradation to be the cause of that failure. That is a whole different chapter. Even here, the CPU have been stable for many and only unstable during certain tasks like UE5 shader compilation, not unstable all around. Even when degradation causes fault, it can be very hard to narrow it down to CPU in many cases, which is why finger has been pointed to everything from UE5 to Nvidia drivers.Except I don't need specialized tools to observe the possibility of operational degradation regarding the CPU's in question, which was my point from the beginning. No goal posts were shifted whatsoever.
In the case of this very thread, Intel CPU's are showing observable degradation due to the fact they crash under operational workloads the CPU's in question should handle with perfect stability - No mystical specialized tools required. This very thread renders your perspective 100% moot.
....Yes you do if you are to make claim your CPU have no degradation. The others observe a point of failure and it points to a degradation to be the cause of that failure. That is a whole different chapter. Even here, the CPU have been stable for many and only unstable during certain tasks like UE5 shader compilation, not unstable all around. Even when degradation causes fault, it can be very hard to narrow it down to CPU in many cases, which is why finger has been pointed to everything from UE5 to Nvidia drivers.
Your statement about having NO observable degradation from OC require you to actually do proper observation of degradation to support that claim. Otherwise its only hot air. You dont have the tools for that and degradation is a slow death, meaning it can be perfectly stable for years until it shorts from degradation or become unstable.
Its not something you should use and can only act like some kind of invalidation of peoples issues or make people think that OC cannot cause degradation of CPUs and other chips for that matter.
My CPU's showed no observable degradation under any workload presented to the processors in question. There was nothing to 'narrow down' as the CPU's in question handled every workload presented to them in a faultless manner. Unlike the Intel CPU's highlighted in the topic of this thread.It's like saying: "My engine handles 500HP at the 7500RPM rev limit it was built to sustain just fine", and then you come along and say: "But after 4000klms you have to run the engine at full load to 8000RPM just to be sure" - Even though the engine still runs fine, makes a perfect 500HP at the 7500RPM limit and is showing no observable faults...
That is factually wrong. In best case, your CPU did not show you any errors in given tasks that you can suspect is caused by degradation. You did no observation of the degradation itself. Even more, besides being factual wrong, that statement does not contribute anything but muddy the issue. I am not trying to correct grammar, that is not my point, my point is that you did not observe any degradation because you dont have the nessesary tools to do so. You have infact no idea the state of the degradation inside your CPU at this stage. In an enthusiast forum, it should be made clear also that OC can reduce the lifespan of your CPU and in some cases it can fail or start to fail within the lifetime of your CPU. Even if you have a stable OC, there is no guaranty that it will remain stable. You might even believe its stable, until you run certain tasks like Cinemabench or UE5 shader compilation as discussed in this thread.....
My CPU's showed no observable degradation under any workload presented to the processors in question. There was nothing to 'narrow down' as the CPU's in question handled every workload presented to them in a faultless manner. Unlike the Intel CPU's highlighted in the topic of this thread.
The only individual invalidating anything here is yourself with talk of mystical CPU degradation dyno's.
Again, how do you know? Did you measure degradation in any way and check if there were any deviation or just blowing hot air about a degradation you most likely have no tools to check for? Are you trying to claim that OC cannot cause more degradation? Or want a pat on your back because your chip had high enough quality to maintain OC so far, while others might have not been so lucky?I've been overclocking since the Celeron 300a days, possibly earlier if you consider I have a Commodore Amiga running an overclocked 68030 25MHz @ 40MHz. I've never encountered any degradation as a result of overclocking in all the years I've been doing it.
No, no it's not.That is factually wrong.
...You can feel free to claim you have OCd a CPU and have no errors that you believe is caused by your OC. But, please dont state that your CPU have no degradation from it without testing it.
The only individual invalidating anything here is yourself with talk of mystical CPU degradation dyno's.
(Coming back to topic)CPU's will be binned at a lower clock speed than the speeds they're realistically capable of to fill a popular slot in the market.
As quoted here:(Coming back to topic)
The suspicion is Intel has binned more aggressively here. Board partners are now increasing voltage & decreasing current. (This can impact all-core boost/top frequencies, effectively converting the chip to a lesser version down the stack)
I totally agree. However, I believe not all CPU's are showing any sign of lasting damage - In a number of cases simply enabling all limiters as per Intel specification seems to resolve any stability issues.So swinging back to the actual topic of this thread: When a CPU runs 1.6v vcore from factory, it's obvious that mobo manufacturers bypassing current limiters by default may result in CPU damage in certain scenarios. Likewise, as someone that's been overclocking for the better part of 30 years without issue, Intel's rated vcore regarding the CPU's in question appears quite extreme - Quite possibly Intel are pushing their silicon the the edge in order to gain an advantage over the competition.
I still think something changed in the microcode which changed core utilization and current and power draw, for certain scenarios. Probably a little after 14th gen launch. Maybe when they added that feature which boosts game performance. Called APO.Early testing of Intel baseline bios by chiphell user:
Raptor Lake stability rate, measured by user kmdkai @ Chiphell
20% stable – 14900K
30% stable – 14900K + Reduced Loadline
40-50% stable – 13900K
50-70% stable – 13900K + Reduced Loadline
60-70% stable – 13900K on B660/B760
100% stable – 13700K
https://www-chiphell-com.translate....html?_x_tr_sl=auto&_x_tr_tl=en&_x_tr_hl=en-GB
https://x.com/3DCenter_org/status/1786026159270224295
If the toothpaste factory continues like this, it will destroy itself. |
It's not different than overclocked RAM or motherboards that claim extremely high OC figures.I am sorry man but the "Failsafe" and "Factory Default" options should not be out of the box OC'd to high hell on any component. You need a safe baseline to start anything from and if Factory Default isn't that then there have been a lot of failures in that product line before you even touched it.
It's very differen't because overclocked RAM isn't overclocked out the gate, you load the auto profile for that CPU and it will start out at 5400 with some loose ass timing even if that RAM is rated for 7000, and you have to implement the timings to move it up to those speeds, sure it will provide you a base "this should work" profile that you can work from there but the base default is always the one the CPU was initially rated for.It's not different than overclocked RAM or motherboards that claim extremely high OC figures.
Everyone thinks it should work out the box but they don't. I wish it just worked also, but they don't lol. It's the name of the game when it comes to anything overclocked. Just because it's printed on the box doesn't mean it's not overclocked so treat it as such and know better. Marketing is marketing they always have crazy claims. It's up to you if you want to deal with overclocked components as a customer.
....what are you talking about?It's not different than overclocked RAM or motherboards that claim extremely high OC figures.
Everyone thinks it should work out the box but they don't. I wish it just worked also, but they don't lol. It's the name of the game when it comes to anything overclocked. Just because it's printed on the box doesn't mean it's not overclocked so treat it as such and know better. Marketing is marketing they always have crazy claims. It's up to you if you want to deal with overclocked components as a customer.
It's very differen't because overclocked RAM isn't overclocked out the gate, you load the auto profile for that CPU and it will start out at 5400 with some loose ass timing even if that RAM is rated for 7000, and you have to implement the timings to move it up to those speeds, sure it will provide you a base "this should work" profile that you can work from there but the base default is always the one the CPU was initially rated for.
....what are you talking about?
I'm not sure what is so hard to understand that parts should be stable out of the box regardless if they're running at the number on the box or not.
Overclocked ram does come stable out of the box. At least the ones I've bought. You usually have to enable a profile to reach the stated overclocked speed.Right but in the case of the Intel CPUs and In
tel motherboards the motherboards are not getting the setting correct in the first place running them unlocked with whatever settings and voltages they see fit. Every motherboard runs different voltages and tweaks it's not universal across any 2 boards so unlocked is meaningless because the board has built in settings for all the different CPUs.
You're delusional if you think overclocked CPUs and motherboards and RAM will work perfectly out the box. You can kick and scream all you want but until the manufacturers do something about it for baseline settings it's just the way it is right now and you have to manually tweak it. Not sure what you don't understand about that?
Your 13900KS is an overclocked processor out of the box, constantly pushing itself to the bleeding edge of performance - You'd want to use those specialized tools to check for degradation. As for overclocked ram, it's called an XMP profile, and I've never seen it enabled 'out the box'.You're delusional if you think overclocked CPUs and motherboards and RAM will work perfectly out the box. You can kick and scream all you want but until the manufacturers do something about it for baseline settings it's just the way it is right now and you have to manually tweak it. Not sure what you don't understand about that?
This actually screwed me upgrading some laptop ram. The bios had no way for using xmp profiles.As for overclocked ram, it's called an XMP profile, and I've never seen it enabled 'out the box'.
Overclocked ram does come stable out of the box. At least the ones I've bought. You usually have to enable a profile to reach the stated overclocked speed.
Nobody is complaining that Intel allows you to go to such high values on their processors. It's the allowing of such potentially damaging settings as factory default that is the problem.
Your 13900KS is an overclocked processor out of the box, constantly pushing itself to the bleeding edge of performance - You'd want to use those specialized tools to check for degradation. As for overclocked ram, it's called an XMP profile, and I've never seen it enabled 'out the box'.
Yes, fast/tight profiles are not guaranteed to work depending on your IMC as well as a number of other factors - Such is the nature of XMP profiles, not too sure just what you're trying to argue here.Absolutely not. RAM has the most instability even with xmp enabled out of the box. A lot of profiles work, then again a lot of profiles don't. A majority of the complaints about RAM is that is doesn't provide perfect stability at XMP settings. Leaving it default is fine but that's not exactly the same as what is being discussed here. RAM has it's own settings.
See above. Linpack and Prime are not enough, you need specialized tools to check for stability/degradation. /sExcept that even my overclocked out the box 13900KS on my high end board with high end RAM 2 different sets I've tried is not stable out of the box. I've had to manually tweak it get it running fully stable in OCCT Linepack and Prime. Not to mention a slew of games.
I've already typed my argument in response to yours. I'm not sure what you're asking me now? I guess I can say the same thing, see above Lol
I've never had a CPU fail on me in the 25 years I've been building my own rigs for a hobby so I am not concerned about it. Every single one of my CPUs have been overclocked mildly. One this is for sure even in my early days I have never left it out the box settings and have always tweaked the settings in the motherboard, that was most of the fun for me having the control in my hands.
- Intel Core i9-13900K "AUTO -253W" - 40/50% (4/5 out of 10 units stable)
- Intel Core i9-13900K "Reduced Loadline" - 50-60% (5/6 out of 10 units stable)
- Intel Core i9-13900K "B760/B660 Board" - 60-70% (6/7 out of 10 units stable)
- Intel Core i9-14900K "AUTO - 253W" - 20% (2 out of 10 units stable)
- Intel Core i9-14900K "Reduced Loadline" - ~30% (3 out of 10 units stable)
- Intel Core i9-14900K "B760/B660 Board" - 40% (4 out of 10 units stable)
And returns. I was in the return window on my 14th gen and let the retailer deal with Intel or their distributor. How many will be resold through those channels?Assuming the i9 buyers are not too numerous compared to the sub $400 cpu crowd. I wonder how many RMA requests Intel is getting over this. Even 10 or 20% of halo buying customers wanting a replacement would be incredibly bad.
yeah they'll prob just modify the internal chip id's and sell them as lower bins to somebody like Dell for non oc'able prebuiltsAnd returns. I was in the return window on my 14th gen and let the retailer deal with Intel or their distributor. How many will be resold through those channels?
Still waiting for Intel. Trust is pretty much gone at this point but May just started.
If you want one, go old school and plan on buying a few to get the silicon lottery roll you expect.
HUB weighs in with a performance and power comparison given the bios updates:
View: https://www.youtube.com/watch?v=bHILyzooR58