Overclock.net › Forums › Intel › Intel Motherboards › [OFFICIAL] ROG Maximus V Owners Club
New Posts  All Forums:Forum Nav:

[OFFICIAL] ROG Maximus V Owners Club - Page 287

post #2861 of 4012
Quote:
Originally Posted by feznz View Post

I think you have answered your own question as everything is single slot configuration

BTW the slots are;
2 x PCIe 3.0/2.0 x16 (x16 or dual x8, red) *2
1 x PCIe 2.0 x4 (black)
1 x mini-PCIe 2.0 x1(top mPCIe Combo expansion card)

I agree with DiGiCiDAL single loop is so much simpler and cheaper weather it is better than multiple loops is debatable but I would say my single loop preforms to my expectations, a separate loop for the VRMs is really not a cost efficient way of doing it.
Use a parallel loop for the GPUs. Overall flow will be higher and CPU temps will be the same for both (or should).
Old rig
(17 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i5 3570K Asus Maximus V Gene MSI Radeon R9 280X Gaming G.Skill Trident 2400 C10 
Hard DriveHard DriveCoolingCooling
Crucial M500 240 Hitach HDT721010 EK Supremacy Clear CSQ Alphacool XT45 360+240 
CoolingCoolingMonitorPower
EK-VGA Supreme HF Swiftech MCP655 Asus MX239H Antec TP 550 New 
CaseAudio
NZXT Switch 810 Black M Creative X-Fi Extreme 
  hide details  
Reply
Old rig
(17 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i5 3570K Asus Maximus V Gene MSI Radeon R9 280X Gaming G.Skill Trident 2400 C10 
Hard DriveHard DriveCoolingCooling
Crucial M500 240 Hitach HDT721010 EK Supremacy Clear CSQ Alphacool XT45 360+240 
CoolingCoolingMonitorPower
EK-VGA Supreme HF Swiftech MCP655 Asus MX239H Antec TP 550 New 
CaseAudio
NZXT Switch 810 Black M Creative X-Fi Extreme 
  hide details  
Reply
post #2862 of 4012
I just finished my simple loop: Warning: Spoiler! (Click to show)
But in planning for my next rig with a much larger one, I was told the same thing from some experienced water cooling guys. Two pumps, one loop is the way to go. Let all your rad space work for you.

As for parallel vs serial for multiple gpus, I have read a lot of differing viewpoints on that from many different people here. Basically it depends on how many gpus, how much pressure you have, and some other specifics of your particular loop. Many people said that with just two gpus it doesn’t really matter which one you use.
Ivy 5.0
(14 items)
 
Ivy 4.8
(15 items)
 
 
CPUMotherboardGraphicsRAM
i7-3770k 5.0 24/7 Maximus V Formula EVGA GTX 690 G.Skill Trident X 2400 (2x8gb) 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Pioneer BD-RW Swiftech H220 Windows 7 Ultimate 64 bit 
MonitorKeyboardPowerCase
2 x NEC 30" Logitech G15 Corsair AX1200 NZXT Switch 810 
MouseAudio
Logitech G9x Logitech G51 
  hide details  
Reply
Ivy 5.0
(14 items)
 
Ivy 4.8
(15 items)
 
 
CPUMotherboardGraphicsRAM
i7-3770k 5.0 24/7 Maximus V Formula EVGA GTX 690 G.Skill Trident X 2400 (2x8gb) 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Pioneer BD-RW Swiftech H220 Windows 7 Ultimate 64 bit 
MonitorKeyboardPowerCase
2 x NEC 30" Logitech G15 Corsair AX1200 NZXT Switch 810 
MouseAudio
Logitech G9x Logitech G51 
  hide details  
Reply
post #2863 of 4012
Quote:
Originally Posted by justanoldman View Post

As for parallel vs serial for multiple gpus, I have read a lot of differing viewpoints on that from many different people here. Basically it depends on how many gpus, how much pressure you have, and some other specifics of your particular loop. Many people said that with just two gpus it doesn’t really matter which one you use.

^ This.

At least in my testing the differences in temperature were nominal, but slightly in favor of serial (GPU temps dropped by ~1C average) and flow increase was nominal at best. I run parallel, but only because I think it looks better, and "all the cool kids are doing it these days" - or something like that.

In all likelihood your most restrictive component in your loop is going to be your CPU block... so that will be the component that sets your flow - no matter how little restriction there is elsewhere. This naturally depends on the GPU block chosen - if you have two or three cards that use fairly restrictive blocks then there might be more of an advantage.

It's similar to the loop order arguments - I run my loop 'wrong' according to many here:

My loop is RES->PUMP->GPU->GPU->CPU->RAD->RAD->RAD->RES

Some will tell you there is an advantage to hitting the CPU block first after the radiators (the theory being that the coolant is at it's coolest and therefore you can OC higher). I disagree with this - despite seeming sensible - since even with my loop set up the way I have it... the deltas are always in the 4-6C range under full load. Although I'm no thermodynamicist (if that's even a thing) to my way of thinking, as long as the coolant is lower than the temperature of the component being cooled - it will absorb as much heat as it's capable of in the time it's directly in contact with the block itself. Since the flow is so significant, that time is limited at best. So unless you're very close to a saturation point of equilibrium (i.e. the coolant is almost as hot as the heatsource), it makes sense - to me at least - that you won't get any better cooling potential of a 60C component with 27C water than you do with 30C water. After all, you are limited to the conductivity of not only the water, but your TIM and the copper of the block itself. If that weren't a limiting factor then it would be possible (without chillers, etc. but with sufficient radiator capacity) to keep a CPU at a fixed temperature regardless of load - and that certainly isn't the case!

The bottom line is that if you want to keep trying different things and tweaking your loops, etc. in the name of "ultimate performance" - then by all means do so! At the end of the day, however, you are likely to have spent hundreds of hours (and potentially thousands of dollars) to extract less than 1-2% greater performance. Performance that could have likely been had by simply paying for a different platform/SKU initially.

It's similar to tuning a Civic... if you do it right (i.e. bigger brakes, reinforced frame & suspension, fuel delivery and ECU, etc - in addition to that big turbo kit) you can get somewhere around 300HP. And you will have spent upwards of $20K in total for a car that's worth $5K, and is still front-wheel drive. You could have just bought a used BMW, 350Z, WRX, etc. in the first place, and not had to lift a finger (other than writing the check that is).

If you love messing with it... then by all means. Otherwise just do what is clean and easy to understand and don't worry about that last little bit - because the LN2 guys will just mop the floor with your scores anyway. biggrin.gif
post #2864 of 4012
Quote:
Originally Posted by DiGiCiDAL View Post

Naturally, your rig = your rules... but as a suggestion, if you put all of those rads into a single loop with everything, you'd most likely get better headroom on everything. Lots of people have tested multiple loops vs single loop configurations and seen no difference or even reduced capacity. I like the look and the concept of multi-loop setups but in all honesty if you got a dual-D5 pumped single loop with the RX360, MCP320 and RS240 combined... you'd probably see a lower delta (mostly because the VRMs don't even really need 120mm of rad cooling on a Z77 board... I don't see anything too extreme and mine are passive).

As an added bonus you'd have redundancy if one of your pumps were to fail... you'd just lose half your flow/pressure, but still be cooling everything... Add to that the fact that if you're not gaming/benching the GPU(s) - you'll have that much more capacity for the CPU... and vice-versa. Food for thought at least. smile.gif
When I added the gpu to my 1st loop on the RX360 my temps definitely rose. Here are my thoughts, I may be wrong. Both cpu and gpu in same loop effect each others temps. Also the more you put in a single loop, more restriction. When I had them in one loop I got some bsod's and crashes. With 2 separate loops, each component has dedicated cooling and does not effect each other. I hadn't gotten any temp related crashes running seperate loops. although I was running a 6990 so the dual gpu's could have a lot to do with it. I'll have to test your theory before assembling final configuration though. You definitely got me thinking! thumb.gif

Off to Microcenter! ( Looks up) Please grant me the serenity to not buy anything that I really don't need!
Edited by hotrod717 - 5/18/13 at 7:33am
    
CPUMotherboardGraphicsRAM
Intel 5820k Asrock x99 OCF V3.1 One of many Kingston HyperX Predator 
Hard DriveOptical DriveCoolingOS
Samsung 840 EVO Asus Bluray Custom Water Win7 Ultimate 
MonitorPowerCaseMouse
Asus VG248QE EVGA P1200 Mountain Mods Extended Custom U2-UFO  Mad Catz R.A.T 5 
Mouse PadAudio
steelseries M-Audio BX5a DLX 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Intel 5820k Asrock x99 OCF V3.1 One of many Kingston HyperX Predator 
Hard DriveOptical DriveCoolingOS
Samsung 840 EVO Asus Bluray Custom Water Win7 Ultimate 
MonitorPowerCaseMouse
Asus VG248QE EVGA P1200 Mountain Mods Extended Custom U2-UFO  Mad Catz R.A.T 5 
Mouse PadAudio
steelseries M-Audio BX5a DLX 
  hide details  
Reply
post #2865 of 4012
Quote:
Originally Posted by hotrod717 View Post

When I added the gpu to my 1st loop on the RX360 my temps definitely rose. Here are my thoughts, I may be wrong. Both cpu and gpu in same loop effect each others temps. Also the more you put in a single loop, more restriction. When I had them in one loop I got some bsod's and crashes. With 2 separate loops, each component has dedicated cooling and does not effect each other. I hadn't gotten any temp related crashes running seperate loops. although I was running a 6990 so the dual gpu's could have a lot to do with it. I'll have to test your theory before assembling final configuration though. You definitely got me thinking! thumb.gif

Off to Microcenter! ( Looks up) Please grant me the serenity to not buy anything that I really don't need!

Well, much of what I stated depends on variables which may or may not be present in any single loop (and most likely involve math and science above my pay grade)... however, since the thermal conductivity of all things involved in the loop are relatively constant... and provided that there is sufficient difference between the temperature of the coolant and the heat source (i.e. greater than the conductivity of the coolant itself at a given flow rate, and capacity of the radiator to dissipate that amount of energy) - it shouldn't really matter.

I perhaps stated my point a little too strongly. It will make a difference in the temperatures of the components if the average coolant temperature rises significantly... however, the point I was making was that the water exiting your GPU was most likely no more than 4-5C hotter than when it entered them... so let's say it entered at 26C and exited at 31C - if it subsequently entered the CPU at 31C and exited at 35C you have a total system delta of 9C - if instead it entered the CPU at 26C it should exit at 30C (roughly) to then enter the GPU at 30C and exit at 5C more... or the same 35C. My guess would be that in your example, the RX360 was simply inadequate (or had inadequate airflow) at dissipating the additional heat overall. I'd still be really surprised if the actual component temperatures were different by more than 1-2C at most.

However, I could definitely be wrong about that... I just haven't seen it myself - and I've read many reviews and experiments that concluded similarly. On the other hand I know people that swear they can hear the differences between power cords on amplifiers... and I can't prove they're wrong - I just know they are! wink.gif
post #2866 of 4012
Quote:
Originally Posted by DiGiCiDAL View Post

Well, much of what I stated depends on variables which may or may not be present in any single loop (and most likely involve math and science above my pay grade)... however, since the thermal conductivity of all things involved in the loop are relatively constant... and provided that there is sufficient difference between the temperature of the coolant and the heat source (i.e. greater than the conductivity of the coolant itself at a given flow rate, and capacity of the radiator to dissipate that amount of energy) - it shouldn't really matter.

I perhaps stated my point a little too strongly. It will make a difference in the temperatures of the components if the average coolant temperature rises significantly... however, the point I was making was that the water exiting your GPU was most likely no more than 4-5C hotter than when it entered them... so let's say it entered at 26C and exited at 31C - if it subsequently entered the CPU at 31C and exited at 35C you have a total system delta of 9C - if instead it entered the CPU at 26C it should exit at 30C (roughly) to then enter the GPU at 30C and exit at 5C more... or the same 35C. My guess would be that in your example, the RX360 was simply inadequate (or had inadequate airflow) at dissipating the additional heat overall. I'd still be really surprised if the actual component temperatures were different by more than 1-2C at most.

However, I could definitely be wrong about that... I just haven't seen it myself - and I've read many reviews and experiments that concluded similarly. On the other hand I know people that swear they can hear the differences between power cords on amplifiers... and I can't prove they're wrong - I just know they are! wink.gif
Lol! I liked that last part. Well, I'm proud owner of a Maximus V Formula. It also seems my prayer wasn't answered as I also bought not only the 3770K, but a 3570k and Raystorm waterblock. Couldn't pass up their advertising error and got the 3570k for $169.99. I also saved $40 on the mobo with bundled deal. I did get Assassin version which helped sweeten the deal. Spent a little more than I wanted, but got everything at once, no waiting on the postman.
It really was a great experience to see all that hardware. Put a lot into perspective, especially the sizes of cases. Wow, that Cosmos II really can't be described with words. Something you have to see to believe. And the Phantom 820 looked a lot better in peson. Just a great experience. I definitely will be going back again. Unfortunately it's about 1 1/2 hours away. At least I have that option though.
What you say makes some sense, however, you can rule out insufficient airfow. I had/have 6 delta wfb1212's in push-pull. I bought a lot of 15 for roughly $2.25 a piece about 8 months ago. Airflow should never be a problem. Noise on the other hand.....haha. Tonight and tomorrow will be sooo fun, except for the fresh os install. However, it does give the opportunity to tweek data placement, organization, and configuration. I will also be installing my ek waterblock on the 7970 Matrix Platinum.
I will be taking some note and photos, but will not be doing a log until it's put in a case. I'll check back in and let you guys know how it goes, temps, ect., ect. thumb.gif
    
CPUMotherboardGraphicsRAM
Intel 5820k Asrock x99 OCF V3.1 One of many Kingston HyperX Predator 
Hard DriveOptical DriveCoolingOS
Samsung 840 EVO Asus Bluray Custom Water Win7 Ultimate 
MonitorPowerCaseMouse
Asus VG248QE EVGA P1200 Mountain Mods Extended Custom U2-UFO  Mad Catz R.A.T 5 
Mouse PadAudio
steelseries M-Audio BX5a DLX 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Intel 5820k Asrock x99 OCF V3.1 One of many Kingston HyperX Predator 
Hard DriveOptical DriveCoolingOS
Samsung 840 EVO Asus Bluray Custom Water Win7 Ultimate 
MonitorPowerCaseMouse
Asus VG248QE EVGA P1200 Mountain Mods Extended Custom U2-UFO  Mad Catz R.A.T 5 
Mouse PadAudio
steelseries M-Audio BX5a DLX 
  hide details  
Reply
post #2867 of 4012
Been a lot of action here.
The way I look at water-cooling it is superior to air BUT the days of huge gains are gone.
even if I managed to lower my temps from lets say 70 degrees to 40 degrees the how much more overclock would that get me?
in my experience maybe 100Mhz on CPU and 70Mhz on GPU
another words I would only tell the difference on benchmarks.

If performance is the aim then spend it on better hardware than water-cooling

Sometimes good marketing will get us to spend more than what we want, then we keep trying to convince ourselves it was all worth it sad-smiley-002.gif
kamikaze
(17 items)
 
  
CPUMotherboardGraphicsGraphics
3770k @ 4.8 asus maximus v extreme Asus Strix 1080 Ti  Asus GTX 770 DCUii 
RAMHard DriveCoolingOS
GSkill 2400Mhz Sandisk ultra pro xspc 240 raystorm Windows 8.1 Pro 
MonitorMonitorMonitorKeyboard
LG UC9938-w AOC 24"  AOC 24" G19 
PowerCaseMouseAudio
enermax maxrevo 1500w nzxt 820 Phantom saitik rat 9 Asus Xonar DX 
  hide details  
Reply
kamikaze
(17 items)
 
  
CPUMotherboardGraphicsGraphics
3770k @ 4.8 asus maximus v extreme Asus Strix 1080 Ti  Asus GTX 770 DCUii 
RAMHard DriveCoolingOS
GSkill 2400Mhz Sandisk ultra pro xspc 240 raystorm Windows 8.1 Pro 
MonitorMonitorMonitorKeyboard
LG UC9938-w AOC 24"  AOC 24" G19 
PowerCaseMouseAudio
enermax maxrevo 1500w nzxt 820 Phantom saitik rat 9 Asus Xonar DX 
  hide details  
Reply
post #2868 of 4012
Quote:
Originally Posted by feznz View Post

Been a lot of action here.
The way I look at water-cooling it is superior to air BUT the days of huge gains are gone.
even if I managed to lower my temps from lets say 70 degrees to 40 degrees the how much more overclock would that get me?
in my experience maybe 100Mhz on CPU and 70Mhz on GPU
another words I would only tell the difference on benchmarks.

If performance is the aim then spend it on better hardware than water-cooling

Sometimes good marketing will get us to spend more than what we want, then we keep trying to convince ourselves it was all worth it sad-smiley-002.gif

Exactly! Marketing is a very powerful tool. It convinces us that there is a reason to spend more money, in order to receive a benefit that in almost every case will either never actually be used or in many cases isn't even logically connected to the purpose of the good/service in the first place. I think of this every time I'm stuck behind someone on the highway that is driving a 6 series BMW... at 55MPH... or a Corvette that's taking a turn at 20MPH... when quite obviously a Toyota Corolla would have done everything they would ever ask of a car for less than a third of the price. wink.gif

I have to admit to tearing my system apart (after less than 3 months) so that I could fit yet another radiator - when I was only seeing 3-4C deltas before... and that only applies during the first 4-5 days when I'm benching and stress testing my OC... I don't even have any time to really play games anymore - but I'm having fun so screw it! biggrin.gif
post #2869 of 4012
Quote:
Originally Posted by DiGiCiDAL View Post

Exactly! Marketing is a very powerful tool. It convinces us that there is a reason to spend more money, in order to receive a benefit that in almost every case will either never actually be used or in many cases isn't even logically connected to the purpose of the good/service in the first place. I think of this every time I'm stuck behind someone on the highway that is driving a 6 series BMW... at 55MPH... or a Corvette that's taking a turn at 20MPH... when quite obviously a Toyota Corolla would have done everything they would ever ask of a car for less than a third of the price. wink.gif

I have to admit to tearing my system apart (after less than 3 months) so that I could fit yet another radiator - when I was only seeing 3-4C deltas before... and that only applies during the first 4-5 days when I'm benching and stress testing my OC... I don't even have any time to really play games anymore - but I'm having fun so screw it! biggrin.gif
You are quite right! But, that's why the term enthusiast applies.

Well, everything went pretty well yesterday with thr install, so far I got 4.4 @ 1.3 stable to some degree. Seems a bit high compared to some that I've seen, but I haven't had that much time with this hardware. This bios is much more complicated than my Saber 990fx and Intel is way different. I really like the board a lot. However the wb seemed harder to install and seems XSPC changed the backplate from what the pictures had shown in the instructions. Previous version was solid and had some holes to replace stock backplate. This one has a altered triangle shape with stand offs and an adhesive. I really had to put a lot of pressure to get the screws started and the springs are about fully compressed without any adjusting.
Still have the wb to install on gpu. Hopefully be able to tear myself away to get that done.
    
CPUMotherboardGraphicsRAM
Intel 5820k Asrock x99 OCF V3.1 One of many Kingston HyperX Predator 
Hard DriveOptical DriveCoolingOS
Samsung 840 EVO Asus Bluray Custom Water Win7 Ultimate 
MonitorPowerCaseMouse
Asus VG248QE EVGA P1200 Mountain Mods Extended Custom U2-UFO  Mad Catz R.A.T 5 
Mouse PadAudio
steelseries M-Audio BX5a DLX 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Intel 5820k Asrock x99 OCF V3.1 One of many Kingston HyperX Predator 
Hard DriveOptical DriveCoolingOS
Samsung 840 EVO Asus Bluray Custom Water Win7 Ultimate 
MonitorPowerCaseMouse
Asus VG248QE EVGA P1200 Mountain Mods Extended Custom U2-UFO  Mad Catz R.A.T 5 
Mouse PadAudio
steelseries M-Audio BX5a DLX 
  hide details  
Reply
post #2870 of 4012
Quote:
Originally Posted by hotrod717 View Post

You are quite right! But, that's why the term enthusiast applies.

Well, everything went pretty well yesterday with thr install, so far I got 4.4 @ 1.3 stable to some degree. Seems a bit high compared to some that I've seen, but I haven't had that much time with this hardware. This bios is much more complicated than my Saber 990fx and Intel is way different. I really like the board a lot...
1.3v is definitely high for 4.4, don’t know if you have been through it but this thread has all the bios settings you need:
http://www.overclock.net/t/1291703/ivy-bridge-overclocking-guide-asus-motherboards

It shows a standard Asus bios, but ours just has more choices. If you don’t already know how to optimize the bios then just copy the settings from that guide, leave any other choices as is.
Ivy 5.0
(14 items)
 
Ivy 4.8
(15 items)
 
 
CPUMotherboardGraphicsRAM
i7-3770k 5.0 24/7 Maximus V Formula EVGA GTX 690 G.Skill Trident X 2400 (2x8gb) 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Pioneer BD-RW Swiftech H220 Windows 7 Ultimate 64 bit 
MonitorKeyboardPowerCase
2 x NEC 30" Logitech G15 Corsair AX1200 NZXT Switch 810 
MouseAudio
Logitech G9x Logitech G51 
  hide details  
Reply
Ivy 5.0
(14 items)
 
Ivy 4.8
(15 items)
 
 
CPUMotherboardGraphicsRAM
i7-3770k 5.0 24/7 Maximus V Formula EVGA GTX 690 G.Skill Trident X 2400 (2x8gb) 
Hard DriveOptical DriveCoolingOS
Samsung 840 Pro 512GB Pioneer BD-RW Swiftech H220 Windows 7 Ultimate 64 bit 
MonitorKeyboardPowerCase
2 x NEC 30" Logitech G15 Corsair AX1200 NZXT Switch 810 
MouseAudio
Logitech G9x Logitech G51 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Intel Motherboards
Overclock.net › Forums › Intel › Intel Motherboards › [OFFICIAL] ROG Maximus V Owners Club