The Cause Of And Fix For Radeon R9 290X And 290 Inconsistency

Page 13 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.


If you were having the same results, wouldn't you wondering the same thing? I think this is the first thought that would come to everyone's mind...
 

fanman

Honorable
Oct 9, 2013
10
0
10,510

Oops, fast you are to response. Fixed already, thanks to quick edit.
 

I think I prefer the enigma of the pre edit original! ;)
 

fanman

Honorable
Oct 9, 2013
10
0
10,510

Feel free to continue the original as well as answer your first question :D
If I bought the card I would switch it to Uber mode forever. For that price I want all the performance I can get. Toms definitely has another idea.
 


But it was AMD who put the switch there in the first place so it would seem that they are ones with the other idea, why else would they put the switch there?
 

anthony8989

Distinguished
Quiet or Uber mode, Shawn? Can you comment on your fan speed? I've read reports that it only takes a 55% fan speed to prevent throttling on the 290x, which is far less than I suspected, given Tom's review.

Sorry I was being unclear. The first question was in reference to your inquiry on Shawn's experience. You were just explaining how you wanted Tom's to provide ambient room temperature in their reviews and yet when it came to Shawn's experience you didn't seem interested in that detail. I meant to point out that the 55% fan speed should prevent throttling in normal room temperatures, but what if the ambient temperature was 10-15% higher than 'normal' - the GPU would again be faced with a throttle.

Any chip with thermal throttling, run above it's thermal threshold, will throttle.

I understand that the GPU is designed to do this, but ~350 mhz is a big throttle-band, wouldn't you say?
 

Darkresurrection

Honorable
Sep 15, 2013
721
0
11,160
when r9-290 and r9-290x were released, Nvidia fans were raged, because they had paid more and were getting less, when the news of inconsistency was released, people started to doubt it, specially owners of r9-290 and r9-290x, that was why they started accusing tomshardware, the reason was that not all cards had this bug, and tomshardware had not stated that, and had not shed light on the problem, when AMD solved the problem by releasing a simple driver Nvidia fans became angry again, because for some days they had started justifying their prior purchase, that was when they started making flame wars they talked about everything from the noise level and talking about the problem's being intact, they are still angry and want to continue this war, if you gave them the source like this one http://www.extremetech.com/gaming/170636-well-that-was-quick-amd-solves-r9-290-throttling-problem-with-a-new-driver , they would say ok, the problem is solved but the card is noisy when you provide them with the source that r9-290 produces a noise level equal to sli 680 and you give links which are published by tomshardware itself, they start going astray, deviating from the topic, and start accusing others of making accounts and double posting, cause they have nothing else to say, that's a shame! and when someone backs AMD up, so that these guys can't mislead people, two to 3 moderators attack that person, and I wont be surprised to see that again, now who is blind, those who don't want to see that the problem is solved, or those who love AMD, because that company cares for them that much to release a card which provides them with the same performance of a card which is twice as expensive!? you can continue buying nvidia, what is the anger for!?
 

Darkresurrection

Honorable
Sep 15, 2013
721
0
11,160

where did the idea of 55% come from? the new driver sets the fan speed at 47% and solves the problem entirely, http://www.extremetech.com/gaming/170636-well-that-was-quick-amd-solves-r9-290-throttling-problem-with-a-new-driver ... 55%!? :lol: now attack me and say dark you are a fanboy...ok I am a fanboy and you are not, what a world :sarcastic:
 

Darkresurrection

Honorable
Sep 15, 2013
721
0
11,160


That's what I am talking about...thanks for confirming my words...they have nothing to say, you see, just say AMD and tomshardware disagrees with you. That's all the big picture you have to see, now if an AMD supporter had posted such a post, they would suspend him, where is the attack the post not the poster policy
 


I think its plain for all to see who is the one going on a mindless rant.
 

Darkresurrection

Honorable
Sep 15, 2013
721
0
11,160


I was posting about the situation and this fact that some guys were mistaken, I had documents, links etc to prove my words, what made you so angry!? reality?
 


Who's angry?
 

Darkresurrection

Honorable
Sep 15, 2013
721
0
11,160

You are, you cant mock at people, because there is a star under your name, did I attack you as a being!? I was replying to that guy who was mistaken, or maybe you, who accuse everyone have 2 accounts!??? If you disagree with me be man enough to say which part of is wrong...what are you a 12 year old!?
 
 

anthony8989

Distinguished
where did the idea of 55% come from?[/b] the new driver sets the fan speed at 47% and solves the problem entirely, http://www.extremetech.com/gaming/170636-well-that-was-quick-amd-solves-r9-290-throttling-problem-with-a-new-driver ... 55%!? :lol: now attack me and say dark you are a fanboy...ok I am a fanboy and you are not, what a world :sarcastic:

We were talking about the R9 290x not the 290. And the idea came from
I've read reports that it only takes a 55% fan speed to prevent throttling on the 290x
- bigpinkdragon286's quest for information. So I'm not really sure why you're asking that question in relation to my post. It's really hard to follow your train of thought. I recommend a proofread before each of your posts.
 

Darkresurrection

Honorable
Sep 15, 2013
721
0
11,160

Pal, your tone of voice was informative, and you were talking about the ambient and your assumptions about throttling, the ambient was not the case and is not the case either, there is no problem with the ambient temperature, the problem was only about fan rotation speed. That 55% was for the uber mode of r9-290x which was set by default, and it didn't have any problem, to provide the user with consistent performance, for the quiet mode of r9-290x 40% was set by default which changes to 47% by default after installing the latest driver, which fully resolves the problem for both cards if it ever existed.
 

Darkresurrection

Honorable
Sep 15, 2013
721
0
11,160
it might be fine to take a look at this http://www.industrialnoisecontrol.com/comparative-noise-examples.htm even 60 dB(A) is considered fairly quiet with r9-290's being arounf 50 dB(A) according to http://www.pcper.com/reviews/Graphics-Cards/AMD-Driver-Updates-Changes-Performance-Fan-Speeds-R9-290X-and-R9-290/Catalyst it shouldn't be a problem for anyone, well it is not for me...the same noise level exists for gtx titan and gtx 780ti http://www.tomshardware.com/reviews/geforce-gtx-780-ti-review-benchmarks,3663-18.html
 

fanman

Honorable
Oct 9, 2013
10
0
10,510


My point of view is from buyers/users. Even AMD put any kind of modes, User/buyer usually choose performance as priority one . Although I'm curious what AMD up to with such low profile mode but I consider that from manufacturer point of view. Just my thought.
 

anthony8989

Distinguished
Pal, your tone of voice was informative, and you were talking about the ambient and your assumptions about throttling, the ambient was not the case and is not the case either, there is no problem with the ambient temperature, the problem was only about fan rotation speed. That 55% was for the uber mode of r9-290x which was set by default, and it didn't have any problem, to provide the user with consistent performance, for the quiet mode of r9-290x 40% was set by default which changes to 47% by default after installing the latest driver, which fully resolves the problemfor both cards if it ever existed.

Dark, if you live in India for example and you don't have air conditioning, and your room temperature is 40 Degrees Celsius, is that 55% fan power going to still prevent throttling? That is my point. And to be honest I wasn't speaking to you in the first place, I really don't know why you're chiming in on this one.

 

Darkresurrection

Honorable
Sep 15, 2013
721
0
11,160


To be honest I don't know about that condition mate, what can I say about it? my room is cool here, even in summer, it is at most 20c-25c... I don't know how I can change the ambient to 40c artificially, specially in this season its a little hard, I will see what I can do...
 
Wow, I'm away a few days and look what I miss!

Anthony, I appreciate your need to keep me held to the same standard I am asking of Tom's. :) Nothing wrong with that, unless you're only motivation is malice, at which point you are only hurting yourself. I think my point is clear that, when reviewing cards and complaining of heat related issues, more details about the situation can be very helpful, that readers can draw better conclusions on their own, and not be made to rely on merely a reviewers conclusion, which may in fact be biased.

If Tom's wasn't concerned with people being able to trust and possibly even separately validate their data, they wouldn't post as many details about their test rigs as they do. When performing tests, you want repeatability, and transparency in the process. While the lack of ambient temperature information may be an oversight up until now, inclusion of the data going forward would be a great step toward repeatability of the testing.

As for my lack of inquiry into Shawn's ambient temperatures, I was simply making polite conversation, rather than attempt a police style inquiry into the every detail of his computer system. I hope that makes sense to you, as I'm not sure how better to explain it. As you can see, he's only posted the one time, so it clearly would have been a further waste of my time to ask that much more.
 

The problem with calling 350 MHz a big throttle band is that it's only your opinion. Unless there are standards defined for what is a big, medium, small, etc. "throttle-band" as you put it, all we can do is argue semantics.

Whether you are happy or not with the amount within which a given chip's frequency may fluctuate, doesn't change the fact that any chip with thermal throttling as part of it's functionality, that is working as intended, and above it's thermal threshold, will throttle.

Can anybody tell me where AMD specifically claims either an R9 290 or 290x in quiet mode will operate under full load without thermal throttling? The fact that there are two modes should be an instant indicator that there are possible reasons for the inclusion of two modes. If you could achieve perfect acoustics and full load with only one mode, I suspect AMD wouldn't have bothered with the extra expense of the switch. It begs the question, who would use Uber mode then, if it was unnecessary? I would expect there to be some sort of compromise from the Quiet mode, but apparently Tom's figures otherwise, and as a result, AMD's updated driver made the Quiet mode compromise that much better, in my opinion, by sacrificing a bit more noise for maintaining a higher upper end clock speed.

In the end, I would never choose to hamstring a performance part in a compromise mode anyway, so Tom's point is mostly lost on me as more of a publicity stunt than a community service.

I still enjoyed reading the article, even if I disagree with parts of it.
 
Status
Not open for further replies.