XsuperiorX schreef op zondag 31 juli 2016 @ 09:58:
Deze komt van sapphire ed
here is a lot of FUD here, where would you like to start?
Before we focus on the cooler, I find it interesting no one has noted that out of the box, the way 90% of consumers will use the card, the Nitro+ has the BEST ability to maintain core boost speed, making it faster than other offerings. However I am guessing since it is not mentioned performance does not matter, lets move to the cooler.
The Nitro+ cooler is capable of maintaining the card, out of the box at it's target of 75C. Now to be fair some sites are reporting higher peak temps but they seem to not be reporting average temps.
For my personal testing I use an open test bench in a room with an ambient temp of around 29C. This is higher than the average ambient temp used by a pretty fair margin, I do this as it is an open bench and want to ensure the temps would give some indication of temp ranges for good air flow cases in cooler rooms.
Next I use GPUz to monitor the video cards core speed, temps, fan speeds and load. I now run a game or benchmark that will loop for a MINIMUM of about 90 minutes, in other words a typical gaming session.
At then end of the loop I pull the data GPUz pulled. I remove all data at the end and beginning of the file that are outside the actual run time of the test. I then import the data into a spread sheet to quickly calculate the averages of the performance metrics I am looking to test.
This is a big deal for temperatures especially since it does not look at one shot spikes but rather at temps over the entirety of the game play session and get a feel for what the cards cooling is really doing over the long haul.
Joker convinced me to use Gears of War for this and I also use Metro Last Light. In both cases over a 90 minute+ run the average temps settled in at around 74C (GoW at 74.315 and Metro Last Light at 74.722) Both of these numbers are dead on our target temp, in other words the cooler worked perfect.
I find it amazing that a core temp of 75C is suddenly consider way to warm and people thing they should be concerned. In the consideration of GPU temps these are well within norms and usually on the cool side.
As for the "noise" of the cooler, this one makes me laugh. First lets dismiss the whole higher end fan speed issue which we know is a driver issue as shown with Asus and our cards. AMD will get it resolved.
Under normal gaming loads our card is very quiet. I should know as I have been sitting next to one on an open test bench stress testing it for a couple of weeks. This BTW is comparing it to my custom cooled loop system that has all it's fans at 5volt and barely makes a whisper even under load.
I am not saying it is silent, that would be a lie and I have never lied to any of you. I can say the noise readings we are seeing are not giving a clear picture.
Lets begin with the fact I have yet to see a review show me proper background information on noise determination. They do not show how they achieved the load, do not explain if this is typical fan behavior or an attempt to see what they can make the noise level be. Do they explain the measuring techniques used as well as give you the ambient room noise, the noise floor, of the room they are in. Have they eliminated all other noise sources before measuring? Do they measure right next to the card or at a distance that would equate to a proper distance your ear is likely to be from the source?
These are ALL important considerations as the noise a product produces is only loud if you hear it? The card I have on the open test bench can only be heard by me and others that have visited, when the card is under heavy load and there is no other noise in the room while they sit next to it.
If a fan is on in the room (many rooms in the summer) or the speakers are on so they can hear the game, the cards noise is drowned out. If the card was in a case you would not hear it.
I understand that my opinion on this is biased and will be taken that way. Also understand I have a reputation among the PC gaming community for never shooting BS, I always give facts straight as I see them. I value that reputation a great deal.
I have talked to about 20 people in recent days that got their Nitro+ from various locations around the world. Some have even posted here on Reddit. ALL of them have said the same thing, they were worried about noise from some reviews they red but once using the card they have found it to be quiet.
Now the question is could we have made a better cooler, well of course we could have. We could have put a massive cooler on it and gotten even lower temps and noise. We would have also had to increase the cost of the card.
The competition used a larger cooler and more fans, they should be cooler and quieter. However they are also slower, with our card being the best 480 on the market at keeping close to it's set boost speeds under load.
We made choices to give the end user, the mainstream gamer, the best performance possible in their out of the box experience. We have built a great card that delivers a reasonable balance of noise, cooling and raw performance.
I firmly believe that our card is a great product and am proud of the work we have put into it.