Netgear Nighthawk AX12 Wi-Fi 6 router review
Corsair HS60 PRO Surround Headset Review
MSI Radeon RX 5700 Gaming X review
Team Group Cardea II NVMe 1 TB SSD Review
ASRock X570 Phantom Gaming ITX/TB3 review
Promo: URCDKey Autumn Sale Windows 10 Pro for $11
Patriot P200 1TB SATA3 SSD Review
ASRock Radeon RX 5700 XT Taichi OC+ review
Guru3D Rig of the Month - September 2019
KFA2 GeForce RTX 2070 Super EX Gamer Black review
AMD Ryzen 5 3600X review
Posted by: on: 07/19/2019 01:34 PM [
132 comment(s) ]
We recently reviewed the Ryzen 7 3700X & Ryzen 9 3900X processors from AMD. How about the 3600 series six core parts eh? Well, let's check out the new Ryzen 5 3600X in this review.
Read article
Advertisement
pages 1 2 3 4 5 6
Aura89
Senior Member
Posts: 7616
Senior Member
Posts: 7616
Posted on: 07/25/2019 09:35 PM
amd wasn't enthusiast for years because the market shunned it.
thats how the market works, the companies don't get to decide they are enthusiast,
Again, no. This is not for you to decide, nor anyone else.
If you want to call what is enthusiast for yourself, then by all means, do it, but claiming you get to decide what others are enthusiasts of, and claiming you get to decide what companies label their enthusiast products as, you must have one hell of a large ego to think you get to decide these things.
And if you continue to assert that you do get to say this information, then this conversation is over, as it's pointless to have a discussion with someone who thinks their word is all.
My complaint was that it's too close to the limit of what a CPU should have.If it's higher than 75c then you need to start worrying a bit.I do acknowledge what AMD did here and for stock it's good.Also know that Intel's coolers are way worse, but sitting a few degrees lower than the "dangerous zone" is bad.Sorry for not making myself clear before.
As long as its stable, there shouldn't be an issue. And if you want better cooling, then you can get better cooling. The reality is this "complaint" doesn't make sense because it implies that giving a cooler, that you can use, for a stable system, or not use if you want a better cooling, is somehow a bad thing?
As i said before i just don't understand really what the complaint is here. Intel doesn't provide you with a CPU cooler more often then not, and when it does, it's horribly inadequate, whereas AMD provides relatively decent cooler that will keep your system stable, and if you want better cooling, you can buy better cooling rather then be forced to have to.
To me it just seems like you expect AMD to bring an "amazing" CPU cooler as stock, undercutting CPU heatsink manufacturers which'll piss them off, and providing better cooling when it's not needed for many people.
amd wasn't enthusiast for years because the market shunned it.
thats how the market works, the companies don't get to decide they are enthusiast,
Again, no. This is not for you to decide, nor anyone else.
If you want to call what is enthusiast for yourself, then by all means, do it, but claiming you get to decide what others are enthusiasts of, and claiming you get to decide what companies label their enthusiast products as, you must have one hell of a large ego to think you get to decide these things.
And if you continue to assert that you do get to say this information, then this conversation is over, as it's pointless to have a discussion with someone who thinks their word is all.
My complaint was that it's too close to the limit of what a CPU should have.If it's higher than 75c then you need to start worrying a bit.I do acknowledge what AMD did here and for stock it's good.Also know that Intel's coolers are way worse, but sitting a few degrees lower than the "dangerous zone" is bad.Sorry for not making myself clear before.
As long as its stable, there shouldn't be an issue. And if you want better cooling, then you can get better cooling. The reality is this "complaint" doesn't make sense because it implies that giving a cooler, that you can use, for a stable system, or not use if you want a better cooling, is somehow a bad thing?
As i said before i just don't understand really what the complaint is here. Intel doesn't provide you with a CPU cooler more often then not, and when it does, it's horribly inadequate, whereas AMD provides relatively decent cooler that will keep your system stable, and if you want better cooling, you can buy better cooling rather then be forced to have to.
To me it just seems like you expect AMD to bring an "amazing" CPU cooler as stock, undercutting CPU heatsink manufacturers which'll piss them off, and providing better cooling when it's not needed for many people.
Evildead666
Senior Member
Posts: 1228
Senior Member
Posts: 1228
Posted on: 07/27/2019 01:21 PM
That's just complete bulls#!t, sorry.
The Fx processors had a safe point of 65c and below, with 75c being the point your IMC began to significantly leak voltage and degrade, it seems to be carrying over to Ryzen for some reason but they are rated for 95c or something around that point
That's just complete bulls#!t, sorry.
Astyanax
Senior Member
Posts: 3239
Senior Member
Posts: 3239
Posted on: 07/27/2019 03:00 PM
I'm quoting a former AMD engineer, whats your source for refutation?
not to mention that most boards operating under AMD specified throttle points will begin to do so at 60c.
That's just complete bulls#!t, sorry.
I'm quoting a former AMD engineer, whats your source for refutation?
not to mention that most boards operating under AMD specified throttle points will begin to do so at 60c.
Goiur
Senior Member
Posts: 673
Senior Member
Posts: 673
Posted on: 07/27/2019 06:35 PM
Former AMD engineer, thats a solid source of information right there... but that guy/girl could be my mother for all i know.
Former AMD engineer, thats a solid source of information right there... but that guy/girl could be my mother for all i know.
Astyanax
Senior Member
Posts: 3239
Senior Member
Posts: 3239
Posted on: 07/27/2019 06:39 PM
ok
https://www.amd.com/en/products/specifications/processors
filter by Fx family and scroll left till you see "Max Temps"
its not like Fx safe temps have been discussed for a decade or anything... oh wait.
Former AMD engineer, thats a solid source of information right there... but that guy/girl could be my mother for all i know.
ok
https://www.amd.com/en/products/specifications/processors
filter by Fx family and scroll left till you see "Max Temps"
its not like Fx safe temps have been discussed for a decade or anything... oh wait.
Goiur
Senior Member
Posts: 673
Senior Member
Posts: 673
Posted on: 07/27/2019 09:22 PM
ok
https://www.amd.com/en/products/specifications/processors
filter by Fx family and scroll left till you see "Max Temps"
its not like Fx safe temps have been discussed for a decade or anything... oh wait.
That source may be valid (until further investigation), but my point is... are you sure that the guy posting is a former AMD engineer or you just need to give some made up info in every post you write.
ok
https://www.amd.com/en/products/specifications/processors
filter by Fx family and scroll left till you see "Max Temps"
its not like Fx safe temps have been discussed for a decade or anything... oh wait.
That source may be valid (until further investigation), but my point is... are you sure that the guy posting is a former AMD engineer or you just need to give some made up info in every post you write.
Astyanax
Senior Member
Posts: 3239
Senior Member
Posts: 3239
Posted on: 07/28/2019 04:26 AM
I'm pretty sure Cliff Maier was definitely with AMD
That source may be valid (until further investigation), but my point is... are you sure that the guy posting is a former AMD engineer or you just need to give some made up info in every post you write.
I'm pretty sure Cliff Maier was definitely with AMD
Aura89
Senior Member
Posts: 7616
Senior Member
Posts: 7616
Posted on: 07/28/2019 04:51 AM
Yes and i spoke to Gabe Newell about Half-Life 3 which is coming out in this september.
....
I feel like you're purposely not getting the point, regardless if it's correct or not.
Someone called you out for your information that didn't include your source.
You then stated your source was a "former AMD engineer", which means squat, since you still did not link to where you got this information, but you did provide a link to another forum which states information contradicting the information you posted by having a lower temperature.
You then were called out on this, as you didn't show who or where this information was provided, and it could be anyone and all we have is your word.
So again you come back without your source, again, but provided AMDs direct website listing information that shows you are incorrect.....again. No where on there does it state that, and i'll quote, "The Fx processors had a safe point of 65c and below, with 75c being the point your IMC began to significantly leak voltage and degrade,", in fact it states that the MAX temperatures range anywhere from 57 degrees, to 90 degrees, depending on what specific FX series processor you had. There's not a single 65 or 75 in there. But even their own website doesn't realistically make sense. It states that the FX-8150 had a max-temp of 61 degrees...what does that mean? above 61 degrees it fails? I mean we know that's not right, plenty of reviews showcase that it is, or can, run higher temperatures then that. So what's the "max temperature" even mean on AMDs website?
So then you come back and reply with what i have quoted. You have finally "listed" your source, by name, but have not actually shown your source' information, where Cliff Maier stated this number you stated wasn't YOUR number but instead Cliff Maiers number.
So what are you not getting about what is being asked of you?
If you don't have a link to Cliff Maier, or proof of any kind that he stated your original statement, then say so!
Otherwise it's just going to keep going on and on and on, about the specific question being asked of you, which is "What is your source, link it, and is there a way to state whatever is being linked is actually written by the guy who you are claiming it is coming from?"
If this can't be answered then state it as such so this back and forth can end since you seem completely unwilling to answer the actual question that has been asked of you.
I'm pretty sure Cliff Maier was definitely with AMD
Yes and i spoke to Gabe Newell about Half-Life 3 which is coming out in this september.
....
I feel like you're purposely not getting the point, regardless if it's correct or not.
Someone called you out for your information that didn't include your source.
You then stated your source was a "former AMD engineer", which means squat, since you still did not link to where you got this information, but you did provide a link to another forum which states information contradicting the information you posted by having a lower temperature.
You then were called out on this, as you didn't show who or where this information was provided, and it could be anyone and all we have is your word.
So again you come back without your source, again, but provided AMDs direct website listing information that shows you are incorrect.....again. No where on there does it state that, and i'll quote, "The Fx processors had a safe point of 65c and below, with 75c being the point your IMC began to significantly leak voltage and degrade,", in fact it states that the MAX temperatures range anywhere from 57 degrees, to 90 degrees, depending on what specific FX series processor you had. There's not a single 65 or 75 in there. But even their own website doesn't realistically make sense. It states that the FX-8150 had a max-temp of 61 degrees...what does that mean? above 61 degrees it fails? I mean we know that's not right, plenty of reviews showcase that it is, or can, run higher temperatures then that. So what's the "max temperature" even mean on AMDs website?
So then you come back and reply with what i have quoted. You have finally "listed" your source, by name, but have not actually shown your source' information, where Cliff Maier stated this number you stated wasn't YOUR number but instead Cliff Maiers number.
So what are you not getting about what is being asked of you?
If you don't have a link to Cliff Maier, or proof of any kind that he stated your original statement, then say so!
Otherwise it's just going to keep going on and on and on, about the specific question being asked of you, which is "What is your source, link it, and is there a way to state whatever is being linked is actually written by the guy who you are claiming it is coming from?"
If this can't be answered then state it as such so this back and forth can end since you seem completely unwilling to answer the actual question that has been asked of you.
Astyanax
Senior Member
Posts: 3239
Senior Member
Posts: 3239
Posted on: 07/28/2019 05:03 AM
Yes and i spoke to Gabe Newell about Half-Life 3 which is coming out in this september.
....
I feel like you're purposely not getting the point, regardless if it's correct or not.
Someone called you out for your information that didn't include your source.
You then stated your source was a "former AMD engineer", which means squat, since you still did not link to where you got this information, but you did provide a link to another forum which states information contradicting the information you posted by having a lower temperature.
You then were called out on this, as you didn't show who or where this information was provided, and it could be anyone and all we have is your word.
So again you come back without your source, again, but provided AMDs direct website listing information that shows you are incorrect.....again. No where on there does it state that, and i'll quote, "The Fx processors had a safe point of 65c and below, with 75c being the point your IMC began to significantly leak voltage and degrade,", in fact it states that the MAX temperatures range anywhere from 57 degrees, to 90 degrees, depending on what specific FX series processor you had. There's not a single 65 or 75 in there. But even their own website doesn't realistically make sense. It states that the FX-8150 had a max-temp of 61 degrees...what does that mean? above 61 degrees it fails? I mean we know that's not right, plenty of reviews showcase that it is, or can, run higher temperatures then that. So what's the "max temperature" even mean on AMDs website?
So then you come back and reply with what i have quoted. You have finally "listed" your source, by name, but have not actually shown your source' information, where Cliff Maier stated this number you stated wasn't YOUR number but instead Cliff Maiers number.
So what are you not getting about what is being asked of you?
If you don't have a link to Cliff Maier, or proof of any kind that he stated your original statement, then say so!
Otherwise it's just going to keep going on and on and on, about the specific question being asked of you, which is "What is your source, link it, and is there a way to state whatever is being linked is actually written by the guy who you are claiming it is coming from?"
If this can't be answered then state it as such so this back and forth can end since you seem completely unwilling to answer the actual question that has been asked of you.
Sometimes the sources can't be linked (email exchange) or is a recollection from a thread on a defunct forum,
or the source was a reddit post whose user deleted their account
Regardless, step off - the information was legitimate, Fx processors can enter thermal runaway at their voltages if they exceed their max temp and decay ensues.
MrInfinit3 as a former contracted employee frequently brings up the imc decay temp, ask him about it. Your hunts are boring.
Yes and i spoke to Gabe Newell about Half-Life 3 which is coming out in this september.
....
I feel like you're purposely not getting the point, regardless if it's correct or not.
Someone called you out for your information that didn't include your source.
You then stated your source was a "former AMD engineer", which means squat, since you still did not link to where you got this information, but you did provide a link to another forum which states information contradicting the information you posted by having a lower temperature.
You then were called out on this, as you didn't show who or where this information was provided, and it could be anyone and all we have is your word.
So again you come back without your source, again, but provided AMDs direct website listing information that shows you are incorrect.....again. No where on there does it state that, and i'll quote, "The Fx processors had a safe point of 65c and below, with 75c being the point your IMC began to significantly leak voltage and degrade,", in fact it states that the MAX temperatures range anywhere from 57 degrees, to 90 degrees, depending on what specific FX series processor you had. There's not a single 65 or 75 in there. But even their own website doesn't realistically make sense. It states that the FX-8150 had a max-temp of 61 degrees...what does that mean? above 61 degrees it fails? I mean we know that's not right, plenty of reviews showcase that it is, or can, run higher temperatures then that. So what's the "max temperature" even mean on AMDs website?
So then you come back and reply with what i have quoted. You have finally "listed" your source, by name, but have not actually shown your source' information, where Cliff Maier stated this number you stated wasn't YOUR number but instead Cliff Maiers number.
So what are you not getting about what is being asked of you?
If you don't have a link to Cliff Maier, or proof of any kind that he stated your original statement, then say so!
Otherwise it's just going to keep going on and on and on, about the specific question being asked of you, which is "What is your source, link it, and is there a way to state whatever is being linked is actually written by the guy who you are claiming it is coming from?"
If this can't be answered then state it as such so this back and forth can end since you seem completely unwilling to answer the actual question that has been asked of you.
Sometimes the sources can't be linked (email exchange) or is a recollection from a thread on a defunct forum,
or the source was a reddit post whose user deleted their account
Regardless, step off - the information was legitimate, Fx processors can enter thermal runaway at their voltages if they exceed their max temp and decay ensues.
MrInfinit3 as a former contracted employee frequently brings up the imc decay temp, ask him about it. Your hunts are boring.
Fox2232
Senior Member
Posts: 9731
Senior Member
Posts: 9731
Posted on: 07/28/2019 08:50 AM
Support your claim with actual evidence, please. Like benchmarks of Zen2 CPU with 5700 XT on PCIe 4.0 against PCIe 3.0.
Wait, Even 2080Ti is on PCIe 3.0.
As for the chipsets comparison insanity. Cheapest to cheapest Z370 is 60% more costly than B450. And on upper limits, Z370 has boards 2.5 times as costly as most expensive B450.
What are you even proposing here? Price comparison for Z370 should be to X470.
And if you go to H370 vs. B450, there is CPU selection. Zen2 offers 65W chips that will not exceed this limit at max clock. While intel offers comparable chips where 65W means base clock on all cores, not turbo. So, you'll have to pay more for H370 just to get most of your CPU.
Smartest choice for H370 could be i7-8700(non-K) which is 6C/12T 65W of intel's TDP and 3.2 ~ 4.6GHz.
And at same CPU price you'll get Zen2 3700X which is 8C/16T 65W AMD's TDP and 3.6 ~ 4.4GHz (higher IPC)... therefore good 25~35% total performance of chip.
And if you enforced AMD's power draw limits to i7-8700, it would be sorry look at performance.
It's actually is. Shorter support term and AMD already stated they will switch off ability to use Zen 2 PCIe lanes in PCIe 4.0 mode on all previous generation motherboards. And there are simply no decent B450 motherboards. They all having poor set of controllers. And if thinking that way, then we can go with Z370 or H370 etc. Obviously previous generations are cheaper. Yet X570 is not enthusiast. It's current gen mainstream. Enthusiast are HEDT, which are Threadrippers and Core-X.
Support your claim with actual evidence, please. Like benchmarks of Zen2 CPU with 5700 XT on PCIe 4.0 against PCIe 3.0.
Wait, Even 2080Ti is on PCIe 3.0.
As for the chipsets comparison insanity. Cheapest to cheapest Z370 is 60% more costly than B450. And on upper limits, Z370 has boards 2.5 times as costly as most expensive B450.
What are you even proposing here? Price comparison for Z370 should be to X470.
And if you go to H370 vs. B450, there is CPU selection. Zen2 offers 65W chips that will not exceed this limit at max clock. While intel offers comparable chips where 65W means base clock on all cores, not turbo. So, you'll have to pay more for H370 just to get most of your CPU.
Smartest choice for H370 could be i7-8700(non-K) which is 6C/12T 65W of intel's TDP and 3.2 ~ 4.6GHz.
And at same CPU price you'll get Zen2 3700X which is 8C/16T 65W AMD's TDP and 3.6 ~ 4.4GHz (higher IPC)... therefore good 25~35% total performance of chip.
And if you enforced AMD's power draw limits to i7-8700, it would be sorry look at performance.
Evildead666
Senior Member
Posts: 1228
Senior Member
Posts: 1228
Posted on: 07/28/2019 11:12 AM
Sometimes the sources can't be linked (email exchange) or is a recollection from a thread on a defunct forum,
or the source was a reddit post whose user deleted their account
Regardless, step off - the information was legitimate, Fx processors can enter thermal runaway at their voltages if they exceed their max temp and decay ensues.
MrInfinit3 as a former contracted employee frequently brings up the imc decay temp, ask him about it. Your hunts are boring.
https://www.amd.com/en/products/cpu/fx-9590
I'll give you that something does indeed look fishy with the FX series "Max Temps" values on AMD's site, and elsewhere.
Thats the sort of link you need to backup your initial post, so you don't get this sort of reply.
(How can you have a max temp of 57°C on a 220W chip ?) I'll be reading in to this some more.
Its does seem related to the 32nm SOI implementation though, so where you imply the correlation with it carrying over to 7nm is a tad baiting.
There have been many process and node changes inbetween then and now, and for this part, i stand by my Bullsh!t statement.
Sometimes the sources can't be linked (email exchange) or is a recollection from a thread on a defunct forum,
or the source was a reddit post whose user deleted their account
Regardless, step off - the information was legitimate, Fx processors can enter thermal runaway at their voltages if they exceed their max temp and decay ensues.
MrInfinit3 as a former contracted employee frequently brings up the imc decay temp, ask him about it. Your hunts are boring.
https://www.amd.com/en/products/cpu/fx-9590
I'll give you that something does indeed look fishy with the FX series "Max Temps" values on AMD's site, and elsewhere.
Thats the sort of link you need to backup your initial post, so you don't get this sort of reply.
(How can you have a max temp of 57°C on a 220W chip ?) I'll be reading in to this some more.
Its does seem related to the 32nm SOI implementation though, so where you imply the correlation with it carrying over to 7nm is a tad baiting.
There have been many process and node changes inbetween then and now, and for this part, i stand by my Bullsh!t statement.
moo100times
Senior Member
Posts: 107
Senior Member
Posts: 107
Posted on: 07/28/2019 12:36 PM
Episode VI: Return of the Troll (I'm not going to quote more of your drivel). Misrepresentation, trying to present old arguments as new, fake and poor sources, combative language, name calling, general baiting and some blatent lies and fiction. I see you have gone all out for the sequel. AMD must be hitting you where it hurts.
ryzen 3k supporting bios's are unstable as all heck on x370 and 470 and this affects all processors. (currently).
Episode VI: Return of the Troll (I'm not going to quote more of your drivel). Misrepresentation, trying to present old arguments as new, fake and poor sources, combative language, name calling, general baiting and some blatent lies and fiction. I see you have gone all out for the sequel. AMD must be hitting you where it hurts.
Astyanax
Senior Member
Posts: 3239
Senior Member
Posts: 3239
Posted on: 07/29/2019 05:55 AM
Episode VI: Return of the Troll (I'm not going to quote more of your drivel). Misrepresentation, trying to present old arguments as new, fake and poor sources, combative language, name calling, general baiting and some blatent lies and fiction. I see you have gone all out for the sequel. AMD must be hitting you where it hurts.
you're cute, ignorant and wrong but cute.
In other news, i just made this thread https://forums.guru3d.com/threads/destiny-2-fix-for-ryzen-3k-in-the-form-of-a-beta-chipset-driver.427800/
the ryzen balanced profile in this chipset driver is much better than the current release driver, give it a try with your ryzen 3k.
https://www.amd.com/en/products/cpu/fx-9590
I'll give you that something does indeed look fishy with the FX series "Max Temps" values on AMD's site, and elsewhere.
Thats the sort of link you need to backup your initial post, so you don't get this sort of reply.
(How can you have a max temp of 57°C on a 220W chip ?) I'll be reading in to this some more.
Its does seem related to the 32nm SOI implementation though, so where you imply the correlation with it carrying over to 7nm is a tad baiting.
There have been many process and node changes inbetween then and now, and for this part, i stand by my Bullsh!t statement.
Few years ago I would have took this as being because the higher voltages on the 9xxx chips would enter thermal runaway faster but for that you'd need uncontrolled cooling which nobody would be doing with a 220w processor.
Rather its the fact that Voltage and Heat together increase the rate in which electrons can hop across oxide barriers, (ie, Quantum tunneling), and this can mean a you get a 1 instead of a 0 (and a bugcheck) in normal cases, or actual damage to a part that is sensitive to rogue electrons in the worst.
in the case of the Fx series, rogue electrons could easily damage the IMC, in the Ryzen 3k series amd have allowed for the chiplets to use more voltage to sustain higher boosts whilst reducing heat via the microprocess change whilst allowing the chip to heat up more.
by splitting the IO components from the cpu cores, they have removed a worst case scenario of a core leaking an electrode into the IMC and destroying it.
Episode VI: Return of the Troll (I'm not going to quote more of your drivel). Misrepresentation, trying to present old arguments as new, fake and poor sources, combative language, name calling, general baiting and some blatent lies and fiction. I see you have gone all out for the sequel. AMD must be hitting you where it hurts.
you're cute, ignorant and wrong but cute.
In other news, i just made this thread https://forums.guru3d.com/threads/destiny-2-fix-for-ryzen-3k-in-the-form-of-a-beta-chipset-driver.427800/
the ryzen balanced profile in this chipset driver is much better than the current release driver, give it a try with your ryzen 3k.
https://www.amd.com/en/products/cpu/fx-9590
I'll give you that something does indeed look fishy with the FX series "Max Temps" values on AMD's site, and elsewhere.
Thats the sort of link you need to backup your initial post, so you don't get this sort of reply.
(How can you have a max temp of 57°C on a 220W chip ?) I'll be reading in to this some more.
Its does seem related to the 32nm SOI implementation though, so where you imply the correlation with it carrying over to 7nm is a tad baiting.
There have been many process and node changes inbetween then and now, and for this part, i stand by my Bullsh!t statement.
Few years ago I would have took this as being because the higher voltages on the 9xxx chips would enter thermal runaway faster but for that you'd need uncontrolled cooling which nobody would be doing with a 220w processor.
Rather its the fact that Voltage and Heat together increase the rate in which electrons can hop across oxide barriers, (ie, Quantum tunneling), and this can mean a you get a 1 instead of a 0 (and a bugcheck) in normal cases, or actual damage to a part that is sensitive to rogue electrons in the worst.
in the case of the Fx series, rogue electrons could easily damage the IMC, in the Ryzen 3k series amd have allowed for the chiplets to use more voltage to sustain higher boosts whilst reducing heat via the microprocess change whilst allowing the chip to heat up more.
by splitting the IO components from the cpu cores, they have removed a worst case scenario of a core leaking an electrode into the IMC and destroying it.
Evildead666
Senior Member
Posts: 1228
Senior Member
Posts: 1228
Posted on: 07/29/2019 12:48 PM
you're cute, ignorant and wrong but cute.
In other news, i just made this thread https://forums.guru3d.com/threads/destiny-2-fix-for-ryzen-3k-in-the-form-of-a-beta-chipset-driver.427800/
the ryzen balanced profile in this chipset driver is much better than the current release driver, give it a try with your ryzen 3k.
Few years ago I would have took this as being because the higher voltages on the 9xxx chips would enter thermal runaway faster but for that you'd need uncontrolled cooling which nobody would be doing with a 220w processor.
Rather its the fact that Voltage and Heat together increase the rate in which electrons can hop across oxide barriers, (ie, Quantum tunneling), and this can mean a you get a 1 instead of a 0 (and a bugcheck) in normal cases, or actual damage to a part that is sensitive to rogue electrons in the worst.
in the case of the Fx series, rogue electrons could easily damage the IMC, in the Ryzen 3k series amd have allowed for the chiplets to use more voltage to sustain higher boosts whilst reducing heat via the microprocess change whilst allowing the chip to heat up more.
by splitting the IO components from the cpu cores, they have removed a worst case scenario of a core leaking an electrode into the IMC and destroying it.
This is getting way off topic, but I think the FX series was probably due to a leaky implementation of the SOI process on 32nm.
It looks like the 32nm was working "ok" and then they decided to dope it with SOI, and it didn't work out so well.
From what I can tell, up to the "Max Temp", the chip should work just fine.
Once you go over the Max temp, leakage starts, and starts bad.
From then, you just HAVE to add more voltage, and generate more heat, to get any more clocks.
I wouldn't expect any damage to the chip unless voltages or temperatures were exceeded though, and I don't mean the 57°C Max temps, but the usual 90/100°C Temps.
You don't tend to get electron migration until you really put high voltage through it, and push higher clocks.
You won't get electron migration or any damage at/around 57°C, that's 100% guaranteed. You couldn't sell a chip that got damage that early.
(I haven't delved deeply into the history if this yet, there must be loads on the internet forums about it, OC.UK etc...).
you're cute, ignorant and wrong but cute.
In other news, i just made this thread https://forums.guru3d.com/threads/destiny-2-fix-for-ryzen-3k-in-the-form-of-a-beta-chipset-driver.427800/
the ryzen balanced profile in this chipset driver is much better than the current release driver, give it a try with your ryzen 3k.
Few years ago I would have took this as being because the higher voltages on the 9xxx chips would enter thermal runaway faster but for that you'd need uncontrolled cooling which nobody would be doing with a 220w processor.
Rather its the fact that Voltage and Heat together increase the rate in which electrons can hop across oxide barriers, (ie, Quantum tunneling), and this can mean a you get a 1 instead of a 0 (and a bugcheck) in normal cases, or actual damage to a part that is sensitive to rogue electrons in the worst.
in the case of the Fx series, rogue electrons could easily damage the IMC, in the Ryzen 3k series amd have allowed for the chiplets to use more voltage to sustain higher boosts whilst reducing heat via the microprocess change whilst allowing the chip to heat up more.
by splitting the IO components from the cpu cores, they have removed a worst case scenario of a core leaking an electrode into the IMC and destroying it.
This is getting way off topic, but I think the FX series was probably due to a leaky implementation of the SOI process on 32nm.
It looks like the 32nm was working "ok" and then they decided to dope it with SOI, and it didn't work out so well.
From what I can tell, up to the "Max Temp", the chip should work just fine.
Once you go over the Max temp, leakage starts, and starts bad.
From then, you just HAVE to add more voltage, and generate more heat, to get any more clocks.
I wouldn't expect any damage to the chip unless voltages or temperatures were exceeded though, and I don't mean the 57°C Max temps, but the usual 90/100°C Temps.
You don't tend to get electron migration until you really put high voltage through it, and push higher clocks.
You won't get electron migration or any damage at/around 57°C, that's 100% guaranteed. You couldn't sell a chip that got damage that early.
(I haven't delved deeply into the history if this yet, there must be loads on the internet forums about it, OC.UK etc...).
Astyanax
Senior Member
Posts: 3239
Senior Member
Posts: 3239
Posted on: 07/29/2019 01:19 PM
Ah, Electromigration.
with voltages so high you'd need to keep temps low to counter conductors and insulations just being eaten away, where as you can allow higher temps on processors that use lower voltage.
I honestly forget what the IMC ran at on the Fx parts, that was VDDR, and VMEM/DDR is the 1.5 or 1.65 signalling voltage for the dram itself?
This is getting way off topic, but I think the FX series was probably due to a leaky implementation of the SOI process on 32nm.
Ah, Electromigration.
with voltages so high you'd need to keep temps low to counter conductors and insulations just being eaten away, where as you can allow higher temps on processors that use lower voltage.
I honestly forget what the IMC ran at on the Fx parts, that was VDDR, and VMEM/DDR is the 1.5 or 1.65 signalling voltage for the dram itself?
vbetts
Posts: 14670
Posts: 14670
Posted on: 07/29/2019 02:21 PM
Alright, FX talk is over.
Back to Ryzen talk.
Alright, FX talk is over.
Back to Ryzen talk.
Evildead666
Senior Member
Posts: 1228
Senior Member
Posts: 1228
Posted on: 07/29/2019 03:06 PM
For another time/forum
edit : Sorry VBetts, just saw your post.
For another time/forum
edit : Sorry VBetts, just saw your post.
Astyanax
Senior Member
Posts: 3239
Senior Member
Posts: 3239
Posted on: 07/29/2019 03:22 PM
In other topics, theres only about a day left till AMD unveils whatever they are doing with the placeholder topic, be interesting to see benchmarks re done when they're done.
In other topics, theres only about a day left till AMD unveils whatever they are doing with the placeholder topic, be interesting to see benchmarks re done when they're done.
slick3
Senior Member
Posts: 1756
Senior Member
Posts: 1756
Posted on: 07/30/2019 01:50 AM
I'm still on the fence about upgrading to this from an i7 6700k. My main focus is gaming; everything else is optional. Thoughts?
GPU is an RTX 2070.
I'm still on the fence about upgrading to this from an i7 6700k. My main focus is gaming; everything else is optional. Thoughts?
GPU is an RTX 2070.
bballfreak6
Senior Member
Posts: 1609
Senior Member
Posts: 1609
Posted on: 07/30/2019 09:18 AM
I'm still on the fence about upgrading to this from an i7 6700k. My main focus is gaming; everything else is optional. Thoughts?
GPU is an RTX 2070.
Mate I just finished building a brand new 3700X system myself a couple of days ago coming from an older 4770K; from my limited gaming tests so far even with a slower GPU like my 1070TI I am noticing better minimum frame rates in Rise of the Tomb Raider and overall smoother gaming experience in other titles.
I'm still on the fence about upgrading to this from an i7 6700k. My main focus is gaming; everything else is optional. Thoughts?
GPU is an RTX 2070.
Mate I just finished building a brand new 3700X system myself a couple of days ago coming from an older 4770K; from my limited gaming tests so far even with a slower GPU like my 1070TI I am noticing better minimum frame rates in Rise of the Tomb Raider and overall smoother gaming experience in other titles.
Loophole35
Senior Member
Posts: 9337
Senior Member
Posts: 9337
Posted on: 07/30/2019 01:50 PM
I'm still on the fence about upgrading to this from an i7 6700k. My main focus is gaming; everything else is optional. Thoughts?
GPU is an RTX 2070.
More cores will definitely help in newer titles. Though I’d look more at the 3700x as a better future proofing option (the new consoles will be running low TDP 8c/16t Zen2 based CPU’s).
I'm still on the fence about upgrading to this from an i7 6700k. My main focus is gaming; everything else is optional. Thoughts?
GPU is an RTX 2070.
More cores will definitely help in newer titles. Though I’d look more at the 3700x as a better future proofing option (the new consoles will be running low TDP 8c/16t Zen2 based CPU’s).
airbud7
Senior Member
Posts: 7409
Senior Member
Posts: 7409
Posted on: 07/30/2019 07:17 PM
with all that said how long do you think a quad core 7700k @5ghz will last?
finally meets minimum requirements?...."eol"
Edit: in other words a 7700k/2080ti/4k monitor setup.....zero frame drop regardless of processor.
More cores will definitely help in newer titles. Though I’d look more at the 3700x as a better future proofing option (the new consoles will be running low TDP 8c/16t Zen2 based CPU’s).
with all that said how long do you think a quad core 7700k @5ghz will last?
finally meets minimum requirements?...."eol"
Edit: in other words a 7700k/2080ti/4k monitor setup.....zero frame drop regardless of processor.
vbetts
Posts: 14670
Posts: 14670
Posted on: 07/30/2019 07:48 PM
with all that said how long do you think a quad core 7700k @5ghz will last?
finally meets minimum requirements?...."eol"
Edit: in other words a 7700k/2080ti/4k monitor setup.....zero frame drop regardless of processor.
Kaby Lake is still a strong architecture. Might not be the highest core count, but there shouldn't be any issues with performance on 4c/8t for awhile on that platform.
with all that said how long do you think a quad core 7700k @5ghz will last?
finally meets minimum requirements?...."eol"
Edit: in other words a 7700k/2080ti/4k monitor setup.....zero frame drop regardless of processor.
Kaby Lake is still a strong architecture. Might not be the highest core count, but there shouldn't be any issues with performance on 4c/8t for awhile on that platform.
airbud7
Senior Member
Posts: 7409
Senior Member
Posts: 7409
Posted on: 07/30/2019 07:56 PM
I think when PC gamers finally realize that a/any modern processor+ a 2080ti/ +a 4k monitor equals the same result......
Just buy the graphics card and power supply to run it.....Crank It Up.
Kaby Lake is still a strong architecture. Might not be the highest core count, but there shouldn't be any issues with performance on 4c/8t for awhile on that platform.
I think when PC gamers finally realize that a/any modern processor+ a 2080ti/ +a 4k monitor equals the same result......
Just buy the graphics card and power supply to run it.....Crank It Up.
pages 1 2 3 4 5 6
Click here to post a comment for this article on the message forum.
Senior Member
Posts: 3239
The Fx processors had a safe point of 65c and below, with 75c being the point your IMC began to significantly leak voltage and degrade, it seems to be carrying over to Ryzen for some reason but they are rated for 95c or something around that point