Overclock.net › Forums › AMD › AMD - General › VRM on the new AM4 motherboards
New Posts  All Forums:Forum Nav:

VRM on the new AM4 motherboards - Page 105

post #1041 of 2078
Quote:
Originally Posted by TheOldTechGuru View Post

I had respect for you, but you are losing it FAST. Fewer cores require less power and generate less heat. The cores that were disabled were more likely disabled due to defects on the die. The reason is irrelevant. It simply takes less power to do what you did, end of story. To try to claim otherwise is ignorance.

Power has nothing to do with cpu-z....a crappy 4 phase junk micro atx can do a validation just fine...if not better.

Heat has everything to do with it, cooler chip higher clock..

Your losing respect for me? I don't care tbh.

I kiss no companies rear...and i will not mislead people into thinking any product is better than it is.

That earns respect from those that matter.

Im sorry but this is how the cpu-z validation game has always been played.

Its a suicide bench.

Smart people validate on best core or minimal amount of cores at the least...

Im waiting to here your response on the msi ln2 cpuz being beat by cinebench stable speeds however. This excuse should be good....
post #1042 of 2078
Quote:
Originally Posted by yendor View Post

Unfortunate. At least it didn't happen to @Zhany.

Huh glad I didn't go through with buying the Titanium, although I seem to be having some rather bad luck with my builds. My ASRock board is on its way back to newegg due to an issue that cropped up with it. Ended up having to get a new case too, power button decided to break off completely. So I won't be back up and running until Monday.
Mini PC
(14 items)
 
  
CPUMotherboardGraphicsRAM
AMD Ryzen 7 1800x Gigabyte Gaming 3 AB350M ASUS Radeon 290 Direct CU II OC CORSAIR Vengeance LPX 16GB (2 x 8GB) 
Hard DriveOptical DriveCoolingOS
Western Digital Blue None Noctua NH-U12S Windows 10 
PowerCase
Seasonic Flagship Prime Titanium 850 Watt Node 804 
  hide details  
Reply
Mini PC
(14 items)
 
  
CPUMotherboardGraphicsRAM
AMD Ryzen 7 1800x Gigabyte Gaming 3 AB350M ASUS Radeon 290 Direct CU II OC CORSAIR Vengeance LPX 16GB (2 x 8GB) 
Hard DriveOptical DriveCoolingOS
Western Digital Blue None Noctua NH-U12S Windows 10 
PowerCase
Seasonic Flagship Prime Titanium 850 Watt Node 804 
  hide details  
Reply
post #1043 of 2078
Btw i don't hate any company if this guy was worshipping asus he would get the same treatment.

I care about only one thing.

Not misleading end users.
post #1044 of 2078
Quote:
Originally Posted by chew* View Post

Btw i don't hate any company if this guy was worshipping asus he would get the same treatment.

I care about only one thing.

Not misleading end users.

I can agree with that easily smile.gif I have no real brand loyalty, I've had good and bad experiences with pretty much any manufacture it just really seems to depend on the platform. Even the same manufacture might be fantastic on Intel but end up being horrific on AMD.
Mini PC
(14 items)
 
  
CPUMotherboardGraphicsRAM
AMD Ryzen 7 1800x Gigabyte Gaming 3 AB350M ASUS Radeon 290 Direct CU II OC CORSAIR Vengeance LPX 16GB (2 x 8GB) 
Hard DriveOptical DriveCoolingOS
Western Digital Blue None Noctua NH-U12S Windows 10 
PowerCase
Seasonic Flagship Prime Titanium 850 Watt Node 804 
  hide details  
Reply
Mini PC
(14 items)
 
  
CPUMotherboardGraphicsRAM
AMD Ryzen 7 1800x Gigabyte Gaming 3 AB350M ASUS Radeon 290 Direct CU II OC CORSAIR Vengeance LPX 16GB (2 x 8GB) 
Hard DriveOptical DriveCoolingOS
Western Digital Blue None Noctua NH-U12S Windows 10 
PowerCase
Seasonic Flagship Prime Titanium 850 Watt Node 804 
  hide details  
Reply
post #1045 of 2078
Quote:
Originally Posted by chew* View Post

Quote:
Originally Posted by TheOldTechGuru View Post

Not as impressive overclocking *fewer cores*. I was referring to the Ryzen 7 series. Lets compare apples to applies.

Lol seriously? Lower skus are all ryzen 7 failed chips...if anything the are harder to oc...

1400 was an 1800x core at one point that failed...

Grow up and downcore a 1800x

Less tears please. You dont dictate the rules for cpu-z franc at canard pc and hwbot does.

WR fwiw is one core..

Actually you know what ill play along.

Oh well that guy needed water.

Not as impressive as air.

Apples to apples i use your bad analogy against you.

I've had quite a different experience with chips that have disabled cores : http://www.3dmark.com/fs/4020922 or this http://valid.canardpc.com/show_oc.php?id=2467876 2 different 960T's the one on the MSI 790FX GD 70 was on an AIO in 70F - fired it up one evening - forgot it was at that setting , played games , browsed all my normal activities for about 6 hours, got ready to go to bed checked the clocks - ooopsie - 4.6 amazing smile.gif .

Quote:
Originally Posted by chew* View Post

Damn that sux. Skyler seems like a cool level headed dude. We were both complaining about our boards last night no fanboy stuff just honest complaints.

I do obviously find this amusing to however.

Not many straws left to grab at now.

Cpu-z ripped away on ln2 and air

Those awsome overbuilt reinforced kevlar bomb shelter pci slots/build quality gone from the equasion.

Lets go back to that flir imaging and the golden stock volts chip biggrin.gif

I'd love to . IF the Titanium is indeed sending more volts to the cpu than it is reporting, wouldn't the VRM's that everyone seems to think are substandard be running hotter than the other boards do?

IF that is also the case , how does it manage this



or this




IF anyone is willing/able to run their rig with ANY AM4 motherboard cpu combination at the same clockspeeds/corevoltage - same loads/ duration on the stock vrm cooling solution at normal ambient temps so we can compare VRM temps it would be greatly appreciated.




Even my 1800X is golden pushing those volts at that speed under 100% load how does it keep vrm's so cool? 49C? 70F ambient temps - stock heatsinks with a low speed 120 fan directed at the cpu socket area.

No straw grabbin here smile.gif
Ryzen Shine!
(14 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 1800X @ 4125mhz 1.432 Volts MSI X370 Titanium   Fury  G.SKILL TridentZ Series 32GB (4 x 8GB) 288-Pin... 
Hard DriveOptical DriveCoolingOS
Samsung 850 pro Hp 1260 I liquid unobtanium Koolance 480mm radiator, 39... Win 7 HP/Winspy 10 64bit 
MonitorKeyboardPowerCase
24" hp @ 1900x1200 Logitech G19 PC power and cooling 910 watt silencer Thermaltake P5 
MouseMouse Pad
Logitech MX 518 fUnc industries 
  hide details  
Reply
Ryzen Shine!
(14 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 1800X @ 4125mhz 1.432 Volts MSI X370 Titanium   Fury  G.SKILL TridentZ Series 32GB (4 x 8GB) 288-Pin... 
Hard DriveOptical DriveCoolingOS
Samsung 850 pro Hp 1260 I liquid unobtanium Koolance 480mm radiator, 39... Win 7 HP/Winspy 10 64bit 
MonitorKeyboardPowerCase
24" hp @ 1900x1200 Logitech G19 PC power and cooling 910 watt silencer Thermaltake P5 
MouseMouse Pad
Logitech MX 518 fUnc industries 
  hide details  
Reply
post #1046 of 2078
Quote:
Originally Posted by chew* View Post

Power has nothing to do with cpu-z....a crappy 4 phase junk micro atx can do a validation just fine...if not better.

Heat has everything to do with it, cooler chip higher clock..

Your losing respect for me? I don't care tbh.

I kiss no companies rear...and i will not mislead people into thinking any product is better than it is.

That earns respect from those that matter.

Im sorry but this is how the cpu-z validation game has always been played.

Its a suicide bench.

Smart people validate on best core or minimal amount of cores at the least...

Im waiting to here your response on the msi ln2 cpuz being beat by cinebench stable speeds however. This excuse should be good....

What does this have to do with CPU-Z? That is just ONE product that does validation. You claimed a higher clock speed, but with a CPU running fewer cores. Excuse me for not being impressed since it is NOT the same chip and it uses less power which means less heat and less strain on the VRM. I assume you know this and are just being an ass. That, or you really don't know what the hell you are talking about.
post #1047 of 2078
Quote:
Originally Posted by chew* View Post

Btw i don't hate any company if this guy was worshipping asus he would get the same treatment.

I care about only one thing.

Not misleading end users.

Right...and claiming a top OC in a discussion about Ryzen 7 OC when you didn't even use the same chip isn't misleading. Dream on...

I have not mislead about anything. I have pointed out numerous *factual* errors and have not been found to be wrong. Buildzoid is only one example, but there are others. So until someone can present *facts*, these empty opinions are just getting a little pathetic.
post #1048 of 2078
Quote:
Originally Posted by TheOldTechGuru View Post

What does this have to do with CPU-Z? That is just ONE product that does validation. You claimed a higher clock speed, but with a CPU running fewer cores. Excuse me for not being impressed since it is NOT the same chip and it uses less power which means less heat and less strain on the VRM. I assume you know this and are just being an ass. That, or you really don't know what the hell you are talking about.

CPU-Z would be the one product used to validate the board 'with less expensive components' you referenced. It's relatively easy to validate high if that's your only goal.

Fewer cores does not necessarily mean less power consumption. I admit I'm cherrypicking a sample here but it's not isolated.

https://www.bit-tech.net/hardware/2017/04/11/amd-ryzen-5-1600x-review/6

One other reviewer noted that power consumption hinted that the disabled cores still have power running through them, which negates a potentially varying amount of thermal headdroom on the chip, likewise strain on vrm. Add in the fact that you won't get symmetrically disabled cores in each ccx all the time and you may have some interesting times with wild variance between two chips. It could be more challenging without limiting the highest clock obtainable. .

But I didn't go deep into testing methodology. Other things could account for power consumption and sample size is still way too low.

It would take testing to confirm the state of disabled cores, ryzen master can see them. OS does not. Ditto effect of non symmetrically located disabled cores in each ccx. Get yer tin foil hat ready or just admit ti's a possibility , not a certainty, if you want to save time. At the very least it would resemble overclocking another chip with disabled defective cores. If only there had been a few of those and someone had experience with them.
post #1049 of 2078
Quote:
Originally Posted by yendor View Post

CPU-Z would be the one product used to validate the board 'with less expensive components' you referenced. It's relatively easy to validate high if that's your only goal.

Fewer cores does not necessarily mean less power consumption. I admit I'm cherrypicking a sample here but it's not isolated.

https://www.bit-tech.net/hardware/2017/04/11/amd-ryzen-5-1600x-review/6

One other reviewer noted that power consumption hinted that the disabled cores still have power running through them, which negates a potentially varying amount of thermal headdroom on the chip, likewise strain on vrm. Add in the fact that you won't get symmetrically disabled cores in each ccx all the time and you may have some interesting times with wild variance between two chips. It could be more challenging without limiting the highest clock obtainable. .

But I didn't go deep into testing methodology. Other things could account for power consumption and sample size is still way too low.

It would take testing to confirm the state of disabled cores, ryzen master can see them. OS does not. Ditto effect of non symmetrically located disabled cores in each ccx. Get yer tin foil hat ready or just admit ti's a possibility , not a certainty, if you want to save time. At the very least it would resemble overclocking another chip with disabled defective cores. If only there had been a few of those and someone had experience with them.

If it truly was fewer cores based on the same architecture it would mean less power. What you are showing is that the cores do not appear to be truly disabled but rather made non-functioning.

That is odd. I can somewhat understand at idle given the common cache architecture and infinity Fabric, but would be expecting a greater difference under load. In the past AMD used a laser to disable cores (after even earlier methods were easily circumvented). It would seem unusual for them to logically disable cores, unless something in the architecture (maybe related to the intelli sense, or whatever they call it) can't be easily segmented and disabled...

There are other sites showing similar results, so that may not be that cherry picked.

But there is a difference under load, just not nearly as much as should be expected if the cores were truly/completely disabled.
post #1050 of 2078
Quote:
Originally Posted by cssorkinman View Post

I've had quite a different experience with chips that have disabled cores : http://www.3dmark.com/fs/4020922 or this http://valid.canardpc.com/show_oc.php?id=2467876 2 different 960T's the one on the MSI 790FX GD 70 was on an AIO in 70F - fired it up one evening - forgot it was at that setting , played games , browsed all my normal activities for about 6 hours, got ready to go to bed checked the clocks - ooopsie - 4.6 amazing smile.gif .
I'd love to . IF the Titanium is indeed sending more volts to the cpu than it is reporting, wouldn't the VRM's that everyone seems to think are substandard be running hotter than the other boards do?

IF that is also the case , how does it manage this
Warning: Spoiler! (Click to show)


or this



IF anyone is willing/able to run their rig with ANY AM4 motherboard cpu combination at the same clockspeeds/corevoltage - same loads/ duration on the stock vrm cooling solution at normal ambient temps so we can compare VRM temps it would be greatly appreciated.




Even my 1800X is golden pushing those volts at that speed under 100% load how does it keep vrm's so cool? 49C? 70F ambient temps - stock heatsinks with a low speed 120 fan directed at the cpu socket area.

No straw grabbin here smile.gif

We were overdue, sacrificing the necessary creatures to keep your chip and board running no doubt wink.gif

Actually when I see your results I wonder what the volts are at the cpu. not because I think titanium is bad. Software is demonstrably not accurate. Could be overreporting. Then your voltage for clocks is better than you think.

It's not mutually exclusive with the position that the mosfets are 'less expensive'. Isolate them and test vs competitor for electrical loss. Or trust spec sheet. Test thermals without the excellent cooling MSI has given them.

I still think more power has to go into those fets to get the same power out, it does not change the fact that they get the same power out with the caveat that software won't accurately report what's getting to the cpu but that's not the mosfets fault. Nature of the business.

Assume, and I don't, that those are the worst mosfets the planet has ever seen, then it makes the rest of the design stand out more for power efficiency, thermal management. if they're only "average" the rest is still good. Better than what the spec sheet seems to show? Wonderful. Conclusively provable.. If only we knew someone with a titanium. >.>

Still envious of the thermals.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: AMD - General
Overclock.net › Forums › AMD › AMD - General › VRM on the new AM4 motherboards