Yes, Macvidcards makes money from modifying video cards. Does that change the fact that the cMP soundly beats apple's latest? no, not one bit. The test stands as proof that the trash can mac pro is a dead end and that my old classic Mac Pro was a good choice. I just got a Titan X from macvidcards and I am quite happy.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
New Barefeats tests show MacVidCards Titan-X the 5K champion, trounces D700s
- Thread starter MacVidCards
- Start date
- Sort by reaction score
Please watch your language !
This is not about delivering power (W) but enough current.
Radeon HD 5870 - 40A (example)
GeForce GTX 970 - 28A
So you recommend cMP users to draw 56A from two 6-pin PCIe connectors ? Very smart.
Just change the subject...
Aparapi does.
Ok, I'll play. Ditch the Titan Black and install the FirePro W9100 for even better performance.
were did you get this figure (28 Amps)?
voltage (V) times current (A) = power (W)
the GTX 970 draws roughly 12 A according to the specs on the NVIDIA homepage.
I'll watch my language, if you watch the facts.
No, I haven't said that.
You claimed two GTX 970 requires extra PSU. That's false.
I claimed two GTX 970 can be fed from the PSU present in a bog standard Mac Pro, which the two setups in the previously mentioned thread proves.
You claimed that total power consumption would rise to 1200W.
In the thread I've linked to the total power consumption amount to about 400W when the GPUs are under render-load.
Such as using Nvidia drivers for a GTX 570 and only getting one monitor vs Mac OSX built in drivers for two monitors. Yeah, that was easy. Updating Mac OSX that breaks Nividia drivers going to a blank screen. Since no boot screen making it much more difficult to fix.
@linuxcooldude
Nice one!
You take a quote from me, where I refer to a DIFFERENT thread where two people have demonstrated that no modding is required to get two GTX 970 up and running.
Then you take a random quote from THIS thread where somebody have gone down the modding road.
What's up with that? It doesn't disprove what I said. It IS possible to run two GTX 970 without hacks.
AndreeOnline (the username you redacted, to make it seem the "modded" originated from the thread I referred to) is probably not even running GTX 970s, since very few of them actually have 6+8 pins. He might, I don't know.
Seriously. If you can't win the argument without deliberately misconstruing what I say, it's not worth continuing this conversation.
Nice one!
You take a quote from me, where I refer to a DIFFERENT thread where two people have demonstrated that no modding is required to get two GTX 970 up and running.
Then you take a random quote from THIS thread where somebody have gone down the modding road.
What's up with that? It doesn't disprove what I said. It IS possible to run two GTX 970 without hacks.
AndreeOnline (the username you redacted, to make it seem the "modded" originated from the thread I referred to) is probably not even running GTX 970s, since very few of them actually have 6+8 pins. He might, I don't know.
Seriously. If you can't win the argument without deliberately misconstruing what I say, it's not worth continuing this conversation.
These requirements are for a GPU + system containing basic components.
Either way, Mac Pro logic board has two 6pin connectors, and we need 4 for this combo.
Y splitters are not solution for everything.
Each 6-pin provides 75W = 150W in total.
Each PCI-E provides 75W = 300W in total (or 150W on the two slots needed).
In other words:
Each GTX 970 can get 75W from the 6-pin and 75W from the PCI-e slot. A total of 150W. According to specs it has a maximum TDP of 145.
Since 145 is less than 150 the point still stands.
EDIT:
Just to add:
If you use both boosters (the two 6-pin) with a single card, that card can get 225W. (2*75W from 6-pin + 75W from slot).
If you use both boosters with two cards, each card will get 75W from the 6-pin and 75W from the slot. A total rating of 300W for the two cards, which have a combined maximum TDP of 290W.
No idea.
In the thread (https://forums.macrumors.com/threads/1840717/) OP mentions having tested:
"Blender (Cuda) , Luxmark (Open CL) , Black Magic (Drives) and Heaven / Valley"
I imagine both the 6-pin and the PCIE slots having some headroom above the specced 75W.
I don't know, if GPUs normally exceed their specified TDPs. According to the thread, the cards only consumed about 130W under load.
EDITED TO ADD:
Link to machine only using boosters + slots to power 2 GTX 970: https://forums.macrumors.com/posts/20637752/
Last edited:
OK, once again to make nMP look better we need to run tests in Windows? How silly is that.
Please show me a single person who has posted about running a FirePro W9100 in OSX. I doubt you can since I just looked in the drivers and it isn't in there. It uses 67A0 and all I saw was 67B0. Maybe you could add it in, but then you have to run in kext dev mode to get driver to load, bigger hack then installing Nvidia drivers.
So, again, OS X Double Precision King is still a Titan-X.
And a code base that AMD built then offered up for Java development is best thing you can find that uses DP? Please.
I went to ALOT of trouble to keep DP switch working with my flashed Titans. And I really wanted to be able to prove that this was a valuable thing, but I never found a way, other then the DP Benchmark that Barefeats has used. Same with the Fermi Pro cards that had much better DP, there just isn't anything in OS X that gets much use from it.
And it costs $3,000.
And it only works in Windows.
And there certainly isn't a way to run one in nMP.
Maybe you should stop before you dig yourself in any deeper.
----------
I'll find out this weekend.
Last edited:
not exactly, to understand performance gap you have to look at the development of intel and gpu makers, this is 2015 and the x58 platform is still holding its own very well, since amd isn't giving intel any competition they are focusing on energy conservation rather than boosting more power. so in the last 6 years intel isn't really making great strides like it used to since AMD isn't giving them a reason to. my x58 pc with xeon can easily keep up with the new haswell and broadwell cpu in terms of gaming and processing.
gpu on the otherhand, that technology is making fun of moore's law, something new always comes out before years end and its usually 20-30 percent improvement over the last gen.
therefore the biggest mistake apple made is making the gpu not swappable with consumer grade gpu, heavily hindering it's lead on keeping the lead.
I just followed by example and changed the context, like you did.
Don't get me wrong, I'm thrilled someone is keeping the cMP up to snuff, thanks for the effort.
I was merely pointing out that in some user cases, the D700 isn't that bad, especially if you do scientific research, which Maxwell isn't really geared for. These solutions aren't public and dwell inside the labs in different faculties. A recent example could be the Bitcoin farming, where you couldn't find a Radeon 7970 in stock anywhere.
Exactly. I couldn't have phrased that better. People who experienced the computer market from the 90's up to the mid-2000s remember well how the CPUs progressed.
CPU Progress from my Atari ST 1040 bought in 1985 to my Mac Performa 630 bought in 1995 -> ~5'000% performance gain.
CPU Progress from the Mac Performa 630 to my Mac Pro 2009 (Xeon 3540) -> ~1'400% performance gain.
CPU progress from Mac Pro 2009 to the last iMac 5K 2015 with Core i7 4 GHz -> ~191% performance gain.
Difference between today's maxed out Mac Pro 2009 with Xeon W3690/X5690 and last iMac 5K with Core i7 4 GHz -> ~51% single core performance. And this is compared with the best CPU Apple currently offers. Most today's new Mac's have a weaker CPUs.
Another thing: The RAM progress in the last years is also negligible. The difference between DDR3 2133 MHz and DDR3-1333 MHz is less than 2%, and there is 4.5% difference from DDR3 2133 MHz to DDR3-1066 MHz: https://forums.macrumors.com/showthread.php?p=21191745#post21191745
So yes, the x58 platform is far from aging.
Last edited:
Really looking forward to your findings! The more benchmarks the better (would be great to see both single and dual GTX 970 benchmarks).
Also, if possible, make a note of the noise level compared to e.g. a GTX 680, which draws more power than a GTX 970.
I currently run a GTX 680, and it believe the 970 is about 3 times faster than that in OpenCL.
I have a GTX 970 on the way. Considering another one - depending on your findings
If you don't get around to benchmarking much, I'll provide single GTX 970 benchmarks when I receive the card.
Can I see tests between Mac Pro 5.1 and 6.1 using same type of technology? Lets say, MP 5.1 using 4 year old GCN GPU, like HD 7970?
Titan X is few months old. It will wipe everything right now on the market. Its pointless to compare it to computer that is 2 years old, and uses GPU that appeared in the world even further in the past.
Im sure, that if Apple will use next gen GPUs from AMD, will wipe down the cMP from pages of history. But, it will surprise anyone?
Technology gets older, and is replaced with newer, better. Why moan about it, all the time? This thread is ONLY for advertising purpose, rather to proving anything.
However, im glad that people who bought GPUs from MVC are happy with them. It shows, that at least hard work that he put into making it running is not thrown away.
Again, one last time. Stop moaning about how awful for your needs is MP 6.1. Go find another solution. Its been two years now, since its on market. People have enough of it.
The end.
Titan X is few months old. It will wipe everything right now on the market. Its pointless to compare it to computer that is 2 years old, and uses GPU that appeared in the world even further in the past.
Im sure, that if Apple will use next gen GPUs from AMD, will wipe down the cMP from pages of history. But, it will surprise anyone?
Technology gets older, and is replaced with newer, better. Why moan about it, all the time? This thread is ONLY for advertising purpose, rather to proving anything.
However, im glad that people who bought GPUs from MVC are happy with them. It shows, that at least hard work that he put into making it running is not thrown away.
Again, one last time. Stop moaning about how awful for your needs is MP 6.1. Go find another solution. Its been two years now, since its on market. People have enough of it.
The end.
No. And you are right.
If Apple makes a 2015-version of nMP it will most likely beat a Mac Pro from 2009 in most areas.
But in 2016 new mainstream/mid-tier GPUs will become available, that will make the Mac Pro 2009 beat the imaginary 2015 nMP in GPU performance once again.
We are not moaning about technology getting newer and better. We are merely sympathising with people stuck with an nMP, that does not get access to newer and better technology at a reasonable price.
(I'm not affiliated with MVC, I enjoy his insights, but have never tried his products. Also, like MVC, I realize each generation of nMP owners will be left behind with each iteration, which is kinda sad.)
Happens, but you know what? It will be seen more and more in future computers, not only from Apple. Why? Because OEMs will get more money that way, than let people upgrade whenever they want. Its obvious.
Today, Apple in their sandbox allow only to use MP as their most powerful computer. For FCPX, Photoshop, Logic its a Hell of a computer.
Apple does not care for the rest. Is that bad? I dont think so.
Also, maybe we should define a Mac Pro as a workstation, for general purpose? Because IMO thats what it really is. A workstation for everyone.
Who can afford it, of course.
Sure, MVP's taunting is immature and annoying (and where exactly are all these "nMP apologists" he keeps referring to in every thread?)... and he's obviously got his agenda to sell some video cards (as long as his customers are happy and he follows forum rules, good for him)... and he uses a lot of straw-man arguments and circular reasoning... and is often rude and belittling to those who disagree with him... but other than that... it's pretty cool what he and others are doing with these cards. As far as I can tell, they are helping people who want to upgrade the video card on their cMP.
The cMP has limitations and the nMP has limitations... that's just the way it is. If someone wants to soup up their cMP with Titan-X video cards, all the more power to them. Let them have their fun benchmarking it against the nMP. If it provides better game play or more productivity, super. It doesn't take anything away from someone who bought an nMP.
The cMP has limitations and the nMP has limitations... that's just the way it is. If someone wants to soup up their cMP with Titan-X video cards, all the more power to them. Let them have their fun benchmarking it against the nMP. If it provides better game play or more productivity, super. It doesn't take anything away from someone who bought an nMP.
Lets be clear on something. Stop the moaning about how awful is the Mac Pro. MVC is doing great job. It has nothing to do with him.
But again. People should stop moaning about this computer. It is, what it is. And thats an end of it.
EOT. Lets get back to REAL topic of this thread. Which is performance of Titan-X in cMP. Without bashing the nMP.
But again. People should stop moaning about this computer. It is, what it is. And thats an end of it.
EOT. Lets get back to REAL topic of this thread. Which is performance of Titan-X in cMP. Without bashing the nMP.
But lets not forget Nvidia is also a leader in graphics cards, so Intel has more then just AMD to worry about.
Depends on how professionals use their workstations. They usually pick and configure it according to their needs during purchase. Then use it with in a set time frame until they think its time for a new one. I'm not going to say a majority of pro's will constantly upgrade in between the next one they purchase. Some do, some don't.
While you said "Consumer Grade" the nMP is workstation grade graphics. We could debate on its the same thing as consumer and its just drivers and so on.
But workstation graphic cards are on a much longer development cycle then the consumer grade ones. Its always been that way. So to say its hindering its lead on consumer grade graphics cards is not the same thing as their both different cards for different uses.
So by the time a new set of workstation graphic cards eventually come out, the pro may already be looking at the next workstation purchase.
That was one of subjects of the topic. But the poster also compared it with the nMP which was not by accident.
Last edited:
Speechless with admiration for GTX970
In honesty I have been dragging my feet for a simple reason. At some point in recent past I thought I had disconnected some power cables and hadn't.
Result being I ripped the plastic plug off the logic board for one of the PCIE power cables. The pins were still there, but the plug itself gone.
I recently found the plug so today I started 970 power testing.
And I am STUNNED.
I have seen GTX285 and GTX570 draw 115-125 Watts from a plug routinely. And I use these as examples as they are both typically dual-6 cards that are considered "safe" in a Mac Pro.
GTX970s start as "reference" cards built on a GTX980 PCB, just with fewer parts soldered on them. In my opinion these were going to draw more power than the custom cards. My favorite card is the PNY GTX970 as it carries same full complement of 3 @ MDP ports. Many cheaper 970s lose 2 of the DP ports which I think people will regret in the future.
Anyhow, I decided to start testing with the reference card as I assumed it would draw more power.
I have so far been unable to get it to use more than 140-145 Watts.
I set Furmark to 5120x1440 and ran it multiple times. No matter what, the test not only finishes every time, it NEVER draws more than 97 Watts from the cable. (I have a dual 6 pin splitter cable in) Meanwhile, i never saw it grab more than 45 from the slot.
Unheard of !!! So, faster than a GTX570 while drawing 80-100 Watts less power. Incredible !!!
Going to do some additional testing with 2nd 970 in.
Note that the app "Hardware Monitor" has the "Line" and "Boost Line" categories swapped.
Moving on to dual card testing, obviously Frmark is a single card test so doubtful I can get both drawing this much.
And this really will mean that we might be able to use 2 of them internally powered.
Thanks again to Nvidia for continuing to support us despite Apple not caring.
Will start a new thread as this isn't related, but wanted to answer the question.
Prepare to be stunned...the power draw of 274 Watts is everything on Line 1.
In honesty I have been dragging my feet for a simple reason. At some point in recent past I thought I had disconnected some power cables and hadn't.
Result being I ripped the plastic plug off the logic board for one of the PCIE power cables. The pins were still there, but the plug itself gone.
I recently found the plug so today I started 970 power testing.
And I am STUNNED.
I have seen GTX285 and GTX570 draw 115-125 Watts from a plug routinely. And I use these as examples as they are both typically dual-6 cards that are considered "safe" in a Mac Pro.
GTX970s start as "reference" cards built on a GTX980 PCB, just with fewer parts soldered on them. In my opinion these were going to draw more power than the custom cards. My favorite card is the PNY GTX970 as it carries same full complement of 3 @ MDP ports. Many cheaper 970s lose 2 of the DP ports which I think people will regret in the future.
Anyhow, I decided to start testing with the reference card as I assumed it would draw more power.
I have so far been unable to get it to use more than 140-145 Watts.
I set Furmark to 5120x1440 and ran it multiple times. No matter what, the test not only finishes every time, it NEVER draws more than 97 Watts from the cable. (I have a dual 6 pin splitter cable in) Meanwhile, i never saw it grab more than 45 from the slot.
Unheard of !!! So, faster than a GTX570 while drawing 80-100 Watts less power. Incredible !!!
Going to do some additional testing with 2nd 970 in.
Note that the app "Hardware Monitor" has the "Line" and "Boost Line" categories swapped.
Moving on to dual card testing, obviously Frmark is a single card test so doubtful I can get both drawing this much.
And this really will mean that we might be able to use 2 of them internally powered.
Thanks again to Nvidia for continuing to support us despite Apple not caring.
Will start a new thread as this isn't related, but wanted to answer the question.
Prepare to be stunned...the power draw of 274 Watts is everything on Line 1.
