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 135

post #1341 of 1990
most SMT FETs has a big plane on the back side. some will be tied to "Drain", some tied to something else (in 87350 case, it's tied to GND). this plane acted as a heat spreader which spread the heat to the PCB.
the bigger package is not equivalent as bigger heat pad. do mind as the efficiency gets better, the package size could be shrunken.
Pinkybeast
(19 items)
 
Mobility
(6 items)
 
 
CPUMotherboardGraphicsRAM
AMD Ryzen 7 1700X PRIME X370-PRO Sapphire RX 480 Nitro+ 4GB G-Skill F4-2400C15-16GVR 
Hard DriveHard DriveHard DriveHard Drive
Samsung 950 Pro 512GB M.2 Western Digital WD4004FZWX Seagate ST3500418AS Toshiba MK5059GSXP 
Hard DriveOptical DriveCoolingOS
Hitachi HDS721025CLA382 Asus BW-16D1HT Noctua NH-U12P SE2 Microsoft Windows 10 
MonitorKeyboardPowerCase
LG IPS LED 22MP68 generic Corsair CX600 BeQuiet Dark Base 900 Pro 
MouseMouse PadAudio
Logitech G402 Corsair MM300 Parasound Z-DAC v.2 
CPURAMHard DriveOS
AMD A6-5350M Patriot DDR3 1600 Sandisk SSD 64GB Microsoft Windows 10 Pro 
MonitorCase
17" HP Pavilion 17 e011nr 
  hide details  
Reply
Pinkybeast
(19 items)
 
Mobility
(6 items)
 
 
CPUMotherboardGraphicsRAM
AMD Ryzen 7 1700X PRIME X370-PRO Sapphire RX 480 Nitro+ 4GB G-Skill F4-2400C15-16GVR 
Hard DriveHard DriveHard DriveHard Drive
Samsung 950 Pro 512GB M.2 Western Digital WD4004FZWX Seagate ST3500418AS Toshiba MK5059GSXP 
Hard DriveOptical DriveCoolingOS
Hitachi HDS721025CLA382 Asus BW-16D1HT Noctua NH-U12P SE2 Microsoft Windows 10 
MonitorKeyboardPowerCase
LG IPS LED 22MP68 generic Corsair CX600 BeQuiet Dark Base 900 Pro 
MouseMouse PadAudio
Logitech G402 Corsair MM300 Parasound Z-DAC v.2 
CPURAMHard DriveOS
AMD A6-5350M Patriot DDR3 1600 Sandisk SSD 64GB Microsoft Windows 10 Pro 
MonitorCase
17" HP Pavilion 17 e011nr 
  hide details  
Reply
post #1342 of 1990
Quote:
Originally Posted by sakae48 View Post

most SMT FETs has a big plane on the back side. some will be tied to "Drain", some tied to something else (in 87350 case, it's tied to GND). this plane acted as a heat spreader which spread the heat to the PCB.
the bigger package is not equivalent as bigger heat pad. do mind as the efficiency gets better, the package size could be shrunken.

Agreed.

My point is that instead of using packages with higher integration, higher efficiency and lower losses msi opted for more parts with less integration, and to make up for heat accumulated from the losses they've used more parts with better heat dissipation and at the same time used a relatively strong heat-sink.
Since the nikos powerpacks are either low or high side they have a relatively easier time dissipating the lost heat than nexfets, drmosfets and the likes that integrate both high and low side. That's actually the only advantage they have that I can think of.

I took some time to go back a few pages in this thread, and 10 pages back or so there are thermal monitoring results from the vrm of various boards. According to these results, the titanium is not far behind in regards to temperatures compared to the crosshair, meaning that it should be around the same temperature levels as the prime (it should actually be a bit better in theory).
post #1343 of 1990
it's all depends on the engineer's perspective. what approach did he use. it's not simple at all.
having a dual mosfet in a package will consume less time to route the PCB trace. one have smaller capacitance while one has faster rise / fall time. which one better? it depends on what perspective they use. which one the engineers thought to be better.
Pinkybeast
(19 items)
 
Mobility
(6 items)
 
 
CPUMotherboardGraphicsRAM
AMD Ryzen 7 1700X PRIME X370-PRO Sapphire RX 480 Nitro+ 4GB G-Skill F4-2400C15-16GVR 
Hard DriveHard DriveHard DriveHard Drive
Samsung 950 Pro 512GB M.2 Western Digital WD4004FZWX Seagate ST3500418AS Toshiba MK5059GSXP 
Hard DriveOptical DriveCoolingOS
Hitachi HDS721025CLA382 Asus BW-16D1HT Noctua NH-U12P SE2 Microsoft Windows 10 
MonitorKeyboardPowerCase
LG IPS LED 22MP68 generic Corsair CX600 BeQuiet Dark Base 900 Pro 
MouseMouse PadAudio
Logitech G402 Corsair MM300 Parasound Z-DAC v.2 
CPURAMHard DriveOS
AMD A6-5350M Patriot DDR3 1600 Sandisk SSD 64GB Microsoft Windows 10 Pro 
MonitorCase
17" HP Pavilion 17 e011nr 
  hide details  
Reply
Pinkybeast
(19 items)
 
Mobility
(6 items)
 
 
CPUMotherboardGraphicsRAM
AMD Ryzen 7 1700X PRIME X370-PRO Sapphire RX 480 Nitro+ 4GB G-Skill F4-2400C15-16GVR 
Hard DriveHard DriveHard DriveHard Drive
Samsung 950 Pro 512GB M.2 Western Digital WD4004FZWX Seagate ST3500418AS Toshiba MK5059GSXP 
Hard DriveOptical DriveCoolingOS
Hitachi HDS721025CLA382 Asus BW-16D1HT Noctua NH-U12P SE2 Microsoft Windows 10 
MonitorKeyboardPowerCase
LG IPS LED 22MP68 generic Corsair CX600 BeQuiet Dark Base 900 Pro 
MouseMouse PadAudio
Logitech G402 Corsair MM300 Parasound Z-DAC v.2 
CPURAMHard DriveOS
AMD A6-5350M Patriot DDR3 1600 Sandisk SSD 64GB Microsoft Windows 10 Pro 
MonitorCase
17" HP Pavilion 17 e011nr 
  hide details  
Reply
post #1344 of 1990
Precisely. It's not a simple "X part is bad, Y part is good". Lot's of variables taking place.
post #1345 of 1990

The images you posted fall right out the window the moment the processors are pushed beyond stock though.

 

No offense meant, but it seems to me you're just conveniently agreeing with people without really understanding anything of what's going on to shine yourself in better light.

   
AGP bencher
(14 items)
 
CPUMotherboardGraphicsRAM
Ryzen R7 1700 Gigabyte GA-AX370-Gaming 5 Sapphire HD 6950 2GiB 2x8GB KFA2 HOF DDR4-3600 
Hard DriveHard DriveHard DriveHard Drive
Crucial MX100 256GB Seagate 600 Series 240GB Seagate 7200.14 2TB Samsung F3 1TB 
CoolingCoolingCoolingCooling
EKWB Supreme HF XSPC Rasa GPU EK XT360 EK 4.0 
OSMonitorMonitorKeyboard
W10 Pro LG IPS235 LG E2250V KUL ES-87 
PowerCaseMouseAudio
SF Leadex II 650W Lian Li PC-A05NB Logitech G9 Xonar DX 
AudioAudio
SMSL SA-S3+Technics CB-250 Sennheiser HD555 
CPUMotherboardRAMHard Drive
AMD A10-5700 Gigabyte F2A75M-HD2 G.SKILL Ares 2133 CL9 Hitachi 5K750 
Hard DriveCoolingOSMonitor
Momentus .7 200GB Noctua NH-L9a Server 2012 R2 Standard AUO B156HW01 
PowerCaseOther
PicoPSU-80-WI-25V AIO Aluminium Handmade TP-Link Archer Something Something Wi-Fi AC 
CPUCPUCPUMotherboard
Core2Duo E6400 Core2Quad Q6600 Pentium Dual Core E5200 AsRock 4COREDUAL-SATA2 R2.0 
GraphicsRAMHard DriveOptical Drive
A dumpload of ancient AGP cards Kingston Value DDR2-667 CL4 2T @CL3 1T Seagate 160GB 7200.10 LG IDE DVD-ROM 
CoolingCoolingOSMonitor
Ghettomade CPU waterblock 49cc 2stroke engine copper radiator WinXP SP2 32bit ProView 17" 
PowerCase
Tacens Radix V 550W Ghetto aluminium bench 
  hide details  
Reply
   
AGP bencher
(14 items)
 
CPUMotherboardGraphicsRAM
Ryzen R7 1700 Gigabyte GA-AX370-Gaming 5 Sapphire HD 6950 2GiB 2x8GB KFA2 HOF DDR4-3600 
Hard DriveHard DriveHard DriveHard Drive
Crucial MX100 256GB Seagate 600 Series 240GB Seagate 7200.14 2TB Samsung F3 1TB 
CoolingCoolingCoolingCooling
EKWB Supreme HF XSPC Rasa GPU EK XT360 EK 4.0 
OSMonitorMonitorKeyboard
W10 Pro LG IPS235 LG E2250V KUL ES-87 
PowerCaseMouseAudio
SF Leadex II 650W Lian Li PC-A05NB Logitech G9 Xonar DX 
AudioAudio
SMSL SA-S3+Technics CB-250 Sennheiser HD555 
CPUMotherboardRAMHard Drive
AMD A10-5700 Gigabyte F2A75M-HD2 G.SKILL Ares 2133 CL9 Hitachi 5K750 
Hard DriveCoolingOSMonitor
Momentus .7 200GB Noctua NH-L9a Server 2012 R2 Standard AUO B156HW01 
PowerCaseOther
PicoPSU-80-WI-25V AIO Aluminium Handmade TP-Link Archer Something Something Wi-Fi AC 
CPUCPUCPUMotherboard
Core2Duo E6400 Core2Quad Q6600 Pentium Dual Core E5200 AsRock 4COREDUAL-SATA2 R2.0 
GraphicsRAMHard DriveOptical Drive
A dumpload of ancient AGP cards Kingston Value DDR2-667 CL4 2T @CL3 1T Seagate 160GB 7200.10 LG IDE DVD-ROM 
CoolingCoolingOSMonitor
Ghettomade CPU waterblock 49cc 2stroke engine copper radiator WinXP SP2 32bit ProView 17" 
PowerCase
Tacens Radix V 550W Ghetto aluminium bench 
  hide details  
Reply
post #1346 of 1990
I've got a Prime Pro. If you guys want I can take some video of running IBT or OCCT at 1.4V and show what HWiNFO and my IR thermometer says, and touch the heatsink with my hand with no fan blowing on it...if someone with a Titanium wants to do the same thing.
post #1347 of 1990
Anyone have an ideas where the Core current numbers come from in HWInfo? the Amperage number of 173.549 seems nonsensical to me? I'm pretty sure the VRM on my board would have gone into full on meltdown if that much power was actually tying to go through. When I manually calculate the amperage based on wattage and voltage I get 105 Watts CPU Package Power / 1.212 CPU Core Voltage = 86.6 amps which makes a lot more sense to me at least.

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 #1348 of 1990
Quote:
Originally Posted by Artikbot View Post

The images you posted fall right out the window the moment the processors are pushed beyond stock though.

No offense meant, but it seems to me you're just conveniently agreeing with people without really understanding anything of what's going on to shine yourself in better light.

You're throwing me an insult and ask me not to get offended. That makes you the third different person who straight off insults me in three pages.

You know who doesn't know what he is talking about? Let me show some examples:

Right now you said I posted images. I did not post related images, not even linked to them. I said that they were posted 10 pages ago or so.


next:
Quote:
Originally Posted by virpz View Post


So, everybody at this point should be aware of what happens when you have not one but TWO CSD87350Q5D per phase versus the MSI X370 Titanium crapfets.

According to PsyM4a, it is almost 3x1.

AsRock X370 Taichi
6-phase Vcore
Per phase: Two CSD87350Q5D
480A capable

ASUS Crosshair VI Hero
4-phase Vcore
Per phase: Two CSD87350Q5D
320A capable

Biostar Gt7
4- phase Vcore
Per phase: Two IR3550 per phase
480A capable

ASUS Prime X370
6-phase Vcore
Per phase: Two CSD87350Q5D
240A capable

MSI X370 Titanium
6-phase Vcore
Phase: One PK616BA High side, Two PK632BA Low side
252A capable according to that manufacturer http://www.niko-sem.com/in/front/bin/home.phtml


Conversation ended. thumb.gif

Here.
For starters the prime has just one fet per phase. The amperage capability posted is also wrong due to this.
Next is the doubling in phases taking place, doubled or not as long as you have coil, driver, high side fet and low side fet all connected to the pwm, with or without doubler, you have a phase. If you had analog (well, hybrid more likely nowadays) pwm instead of digital, you could have even more sophisticated setups with interleaving allowing something like 10 phases with 3 doublers using a 6 phase pwm or something. The point is that the numbers on that post are messed up.

next:
Quote:
Originally Posted by chew* View Post









Now what was that you were lecturing me about?

Was it moisture?

Sorry to make you look dumb but you do that fine without help it seems...

People who do this on the professional lvl do know what they are doing.

The guy here posts ln2 session photos of years old components and insults me right to my face in the middle of a conversation about completely unrelated components... to prove his point. All that without forgetting to remind the world that he is a "pro".

In short, he's telling me "mine is bigger than yours, so I win".

No comment.

next:
Quote:
Originally Posted by virpz View Post







Boy, oh boy.

You are missing the basics of Ohm's law, but hey. You did you fall into that trap like nobody else.

10/10
thumbsupsmiley.png: keep teaching.gif

The guy here quoted multiple posts from multiple pages back just to tell me that I fell into a trap "like nobody else".... cause you know, this is war and such. Who needs to have an adult conversation when you can just say "I win, you lose, conversation ended" or something.

Quote:
Originally Posted by bardacuda View Post

I've got a Prime Pro. If you guys want I can take some video of running IBT or OCCT at 1.4V and show what HWiNFO and my IR thermometer says, and touch the heatsink with my hand with no fan blowing on it...if someone with a Titanium wants to do the same thing.

You could just tell us. It'll be interesting to see how the board responds to load changes. This is one of the few reasonable posts in the last pages.
Edited by PsyM4n - 5/2/17 at 6:35am
post #1349 of 1990
Seems best place to ir is back of board cuda.

Stick some black tape on hs..ir gets accurate smile.gif

You challenged my abilities and my knowledge psyman...

You said its impossible. Moisture kills!!!

I showed you its not impossible..moisture is my friend.

The fact that i am a pro or not has nothing to do with it. An average guy can do the same...

The fact that they are age old components...lol.

Ok ill take a pic of next session and i will pour water on my board in a video just for you...

Done it before for non believers do it again.no worries...
Edited by chew* - 5/2/17 at 6:42am
post #1350 of 1990
Quote:
Originally Posted by PsyM4n View Post

You're throwing me an insult an ask me not to get offended. That makes you the third different person who straight of insults me in three pages.

You know who doesn't know what he is talking about? Let me show some examples:

Right now you said I posted images. I did not post related images, not even linked to them. I said that they were posted 10 pages ago or so.


next:
Here.
For starters the prime has just one fet per phase. The amperage capability posted is also wrong due to this.
Next is the doubling in phases taking place, doubled or not as long as you have coil, driver, high side fet and low side fet all connected to the pwm, with or without doubler, you have a phase. If you had analog (well, hybrid more likely nowadays) pwm instead of digital, you could have even more sophisticated setups with interleaving allowing something like 10 phases with 3 doublers using a 6 phase pwm or something. The point is that the numbers on that post are messed up.

next:
The guy here posts ln2 session photos of years old components and insults me right to my face in the middle of a conversation about completely unrelated components... to prove his point. All that without forgetting to remind the world that he is a "pro".

In short, he's telling me "mine is bigger than yours, so I win".

No comment.

next:
The guy here quoted multiple posts from multiple pages back just to tell me that I fell into a trap "like nobody else".... cause you know, this is war and such. Who needs to have an adult conversation when you can just say "I win, you lose, conversation ended" or something.
You could just tell us. It'll be interesting to see how the board responds to load changes. This is one of the few reasonable posts in the last pages.

yours included. you are not abolished from "unreasonable posts" because you are now all (seemingly) offended.
those posted images? FLIR? are trash to use as anything but "pretty colors" anyone using them to back up their point - instantly invalidates the point, but you knew that right?


the problem we are having - is that we have too many "armchair" engineers here - and we have people who have been doing this for more years than most of us combined; I wonder who is more correct?
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