Originally Posted by Strider49
Yeah, I know the Classy has a PCB that is 11'' long and that one cannot place the res to the right of the MoBo without some modding. But I'm actually talking about the width of that same PCB. As you can see in the image by Kurodragon in my previous post, the card blocks the top few holes on the back panel of the case Phanteks provided for mounting a res there. Will I still be able to install the Phobya 50mm res there? If not, then I think that's a big flaw in the design of the case.
Well I don't know that I'd call it a flaw ..... in any design you have to make compromises and I think Phanteks went all out to accommodate 95+% of the cards out there.....yes a 1/2 inch or even a few mm here and there would make it accommodate a few more things and this was referenced in my "Enthoo Primo II Suggestions" post, but eventually, when you keep making the case bigger and bigger to accommodate everything, then you are no longer unique .... you basically have a Case Labs case ad a new price to match.
The MSI got a 9.9 on TPU's review, but it has a reference PCB and, as far as I know, the brand only offers a 2 year warranty in my country, unlike Asus and EVGA who offer 3 years. That's why I only consider these two brands. There are two problems with the Asus card however: one is that the warranty is voided as soon as one takes the DirectCU II cooler off; and second, Guru's high precision thermal imaging equipment reported some really high temps on the VRM (~100ºC), and, considering I'm going to run the card on air for the next few months, this is a no-no for me.
It has a reference PCB ? That would be a significant change. The only factory card I know of that has historically had a reference PCB is the EVGA SC series. The MSI and Gigabyte Windforce cards have consistently led the pack in the 7xx series with the highest factory OCs and they have been able to do that by using a multi phase VRM and custom PCB.
If MSI has gone to a reference PCB and VRM, I am a bit puzzled how they can get a card on the market with those kind of clocks matching the Windforce and again embarassing Asus with their measly 954 MHz. I notice that the MSI review says "MSI is using the same OnSemi NCP4206 voltage controller as on the reference design. It supports software voltage control and monitoring via I2C." implying only that it using the same voltage controller as the reference design..... which is still disappointing unless nVidia had decided to ante up on their own for the Ti and put a beefier one in there that renders the upgrades unnecessary.
As to the article and hi VRM temps, that's not uncommon (as indicated by your Asus reference)..... but that will be a non issue with water cooling, my VRMs never break 50C. I do recall.....don't know what review it was but they noted that the VRM was missing a thermal pad which would account for unusually high temps. Perhaps the Asus Poseidon is your answer.... provided great air cooling and water cooling btho I have to say I am not aware of what ya do with the fan shroud once ya move to water and how long the PCB is.
Again, back in the day when I did the "Enthoo II" recommendations post, Phanteks did advise that they were already working on "Reservoir Bracket II", but given the delays on the new cases, accessories and such, I'm not quite sure where it sits on the delivery schedule.
Originally Posted by Leeford
Thanks for that explanation of PWM fan header & VV fans headers. I wonder how the fan headers on this PSU would function with the hub. (This has 4 case fan headers)
Here is the quote from the manual concerning their usage. I guess they are simply more VV headers but not sure if they operate the same as Mobo headers. This seems to use a case heat sensor to determine fan speed where a mobo header uses the board heat, which may or may not be different.
I don't get it..... the PSU in the Enthoo is spacially and thermally isolated to a great extent from the rest of the case.... Where is it's thermal sensor the PSU is using ? If in the PSU, how will it respond to temperatures on the other side of the MoBo backplate ? If you are water cooling, then we certainly wouldn't want case temps controlling rad fans so this would control the front fans, rear fan and optional side panel fans ?
Originally Posted by chrisnyc75
Yeah, I would agree that the software is more user-friendly, and therefore preferable, but the problem is that it's designed as an interface with the bios.... so fan adjustments made in AI Suite II cause changes to the bios settings. But since the bios settings only allow a straight line (i.e. low point and high point) curve for fan control, the third point in the software curve gets lost -- or possibly conflicts with the bios settings. In other words, if I turn OFF fan control in the bios, but turn it ON in AI Suite, when I go back into the bios fan control is turned back ON (by AI Suite).
Thanks for the insight, though. I'm going to play around with it more this weekend and keep an eye on the fans. It may just be a matter of finding the rpm setting that these particular fans are happy with and leaving it at that. My cooling system is so overkill I don't think the fan curve really makes much difference anyway.
I can tell you that FanXpert2 trumps the BIOS. For example..... in BIOS all Fan Q Control is enabled and lists a minimum fan speed of 600 rpm. However what I see in FanXpert2 is:Rad Fans - Text Description:
Max Speed = 1201 rpm
Min Speed = 285 rpm
Fan Run when Power Is > Than = 372 rpm
Controllable Range = 29% - 100%Rad Fans - Graph Description:
0% = 0 rpm
10% = 0 rpm
20% = 0 rpm
30% = 432 rpm
40% = 567 rpm
50% = 694 rpm
60% = 821 rpm
70% = 931 rpm
80% = 1041 rpm
90% = 1149 rpm
100% = 1201 rpm
Measured Fan Speed at 25C = 420 rpm (30% mark on curve)Case Fans - Text Description:
Max Speed = 1289 rpm
Min Speed = 321 rpm
Fan Run when Power Is > Than = 258 rpm
Controllable Range = 26% - 100%Case Fans - Graph Description:
0% = 0 rpm
10% = 0 rpm
20% = 0 rpm
30% = 455 rpm
40% = 605 rpm
50% = 739 rpm
60% = 875 rpm
70% = 991 rpm
80% = 1109 rpm
90% = 1230 rpm
100% = 1239 rpm
Measured Fan Speed at 25C = 383 rpm (28% mark on curve)
So we can see that both sets of fans are spinning slower than the 600 rpm minimum in the BIOS. We also see that % of power does not = % of speed and i dunno why the same fan responds differently whether it's on rads or case tho I expect it's due to SP. When running stress tests, the fans do stay on curve.
1. Leave BIOS set to Enabled / Standard / 600 rpm and then
2. Go to the initial FanXpert2 page (not the individual fan pages) and click the FAN TUNING button on the lower left of the top window.
3. Let it do it's thing, fans will speed up / slow down / turn off / turn on.
4. Set ya curves.
I have for Rad fans at the moment:
30% @ 30C
65% @ 60C
100% @ 75C
and for Case Fans
28% @ 40C
65% @ 60C
100% @ 75C
Originally Posted by Gabrielzm
we are now official
Damn now I need a new case .... don't feel rebellious anymore
Originally Posted by Mr Mari0o
Hey guys I plan on buying one of these cases in white. I'm currently debatng on the rad size for the bottom of the case. I currently have a nexxos xt45 480 for the top and I was thinking of getting a second one for the bottom. I was also Looking at the monsta 80mm 360.
Which would be best dual xt45 480's or a xt45 480+monsta 360?? I only have one 780ti and 4770k to cool at the moment but I plan on throwing a second gpu in there before the end of year.
Any and all help will be very greatly appreciated
For twin 780 Tis you will be able to maintain the desired Delta T of 10C with a 4.6 Ghz OC on a 4770k and 25% OCs on twin 780 Tis (with fan filters removed) with just a 420 + 280 and only fans in push. With stock filters in place and no other bottom mods that goes up to 12.8C...... knock an estimated 21% off those numbers on push pull.
An XT45-480 will give ya just 5 less watts than the 420 in push ..... here's some numbers for comparison:
XT45-420 + UT60-280 = 420 watts in push / 508 in push-pull (Reference)
XT45-480 + XT45-480 = 490 watts in push / 592 in push-pull
XT45-480 + Monsta-360 = 442 watts in push / 535 in push-pull
To download charts summarizing all martins tests for the 360 rads and extrapolations for 120, 240, 480, 140, 280 and 420, go to:
Originally Posted by jrcbandit
So my main questions regarding the Enthoo Primo:
1) Does the Enthoo Primo have the option to use screws normally in the 5.25 bays and can you remove with the door covering the 5.25 bay (and does it look okay to do so)?
2) Will it be easy to fit 4 hard drives and a large radiator on the bottom of the case? I have a 360 radiator at top and I was thinking of upgrading my 120 radiator to something larger - just looking at the case it seems a 240 radiator would absolutely be 0 issue, but I am not sure about a 360 or larger radiator.
To install my temp monitor / fan controller, I had to pop off the latch as the unit is too short. I used screws and then popped the latch back on for aesthetics. The latch holds the optical snugly and w/o vibration. Can't offer an opinion on bat res's w/ pumps tho as I'm not a fan of having that stuff up in there..
Originally Posted by jrcbandit
Thanks for the info, sounds like the case could work well for me. I have 3 normal hard drives that could fit in the bay and a SSD that could go in the special SSD mount. The problem with the ATSC 840 screwless option was that it only has all the screw holes on the right side, it lacked half the screw holes on the left side which so happened to be the ones that the reservoir use/line up with...
With a single D5 vario pump, a CPU and GPU block, would 2 360 radiators be recommended or should I go 360 and 240? Someday I may get a 2nd video card so that would then be two GPU blocks.
Yes, screw holes on both sides. The case is perfectly designed for a 4.6 GHz Haswell and twin top end GFX cards (780 / 780 Tis). From a fitting and usage standpoint, to get max use of all case features (i.e res bracket , easily accessible top fill port, access to all MoBo features and cables, no blocking of cable grommets, etc.) an XT45-420 and UT60-280 will provide all necessary cooling needs for a Delta T of 10.5C. I'll compare that with your listed options
XT45-360 + UT60-240 w/ 1250 rpm fans = 308 watts in push / 374 in push pull
XT45-360 + UT60-360 w/ 1250 rpm fans = 371 watts in push / 449 in push pull
XT45-420 + UT60-280 w/ 1250 rpm fans = 420 watts in push / 508 in push pull
-The 360 and 240 will cost ya $157 and you'll need 5 fans for $90 fans for $75 to get started w/ 1 card or $227 ($0.74 per watt) .... add $75 when ya go to push / pull w/ 2nd GFX card for a $302 total cost
-The 360 and 360 will cost ya $180 and you'll need 6 fans for $90 to get started w/ 1 card or $270 ($0.73 per watt) .... add $90 when ya go to push / pull w/ 2nd GFX card
for a $360 total cost
-The 420 and 280 will cost ya $162 and you'll need 3 more Phanteks fans * for $45 to get started w/ 1 card or $207 ($0.49 per watt) .... add $75 when ya go to push / pull w/ 2nd GFX card for a $282 total cost
Case comes with 5 fans, 2 of which are available for rad usage at top and bottom.
Originally Posted by kangk81
Unfortunately, I'm also new to WC as can be seen by my noobish tubing in my pic 2-3 post up but what I can tell you is a single GPU can run hotter than a single CPU when gaming. So for a single GPU & single CPU config, a single 360mm or 480mm rad should be more than enough. There is no hard and fast rule for WC which is ultimately decided by your budget and imagination. I recommend reading the WC section of OCN for more ideas and guides.
The "rule of thumb" oft quoted .... 1x120mm for each component + 1 spare doesn't rung true for me..... to my mind a rule that uses the same rad allocation for a 100 watt and 300 watt GFX card might be fine in the middle range but has gotta be outta whack at the edges. It's a simple matter to calculate max theoretical load just by adding up the wattage of your components. For example I calculated 792 watts for my system at max OC and running Furmark I draw about 685 from the wall (CPU currently at stock and GFX cards a lil less than max normal OC)....That's about 620 watts "inside the PC" but figure add 30 watts to get GFX back at max OC and I can read from the utilities that my CPU under Furmark is 100 watts under its max as Furmark doesn't push CPU..... so I am at 750 .... perty darn close to the calculated and my pump is only at half speed so that shaved off another 23 watts.
So knowing that's good, all ya have to do is grab ya rad's capabilities and compare the two. Now keeping in mind that lotta heat is directly radiated by items no in martins rad tests (i.e. rad shroud, MoBo heat sinks, backplates, blocks, fittings, tubes etc)..... add in the fact that nothing you are ever going to run will stress everything at the same time.... through trial and error, and asking around for system / temp data, we came up with a factor of 60% to be handled by the rads. So let's say you are water cooling a 4770k (140 watts) and a 780 Ti (300 watts) both overclocked, a MoBo Block (20 watts) and a pump wasting some heat to the loop (20 watts), you'd have 480 watts .... 780 with a 2nd card.
480 x 60% = 288 watts
780 x 60% = 468 watts
An XT45-420 on the top of the Enthoo would give ya 250 watts and a Delta T of about 14.8C in push w/ 1250 rpm fans and 1 GFX card
An XT45-420 on the top and UT60-280 on the bottom of the Enthoo would give ya 420 watts and a Delta T of about 8.8C in push w/ 1250 rpm fans and 1 GFX card
An XT45-420 on the top and UT60-280 on the bottom of the Enthoo would give ya 508 watts and a Delta T of about 11.8C in push-pull w/ 1250 rpm fans and 2 GFX cards
All of Martins test data for 360's is summarized and downloadable in spreadsheet format here with extrapolations for 120. 240, 480, 140, 280 and 420 rads also tabulated.
http://www.overclock.net/t/1457426/radiator-size-estimatorEdited by JackNaylorPE - 4/16/14 at 11:22am