Another question on temperature

AnPei wrote:
GB seal does not make it not news .
Don't lie puts on the screw, it is on windforce. On ГТ650 and paper seal on the side was.

saa0891 wrote:
I understand at least some of the arguments to expect from you is meaningless.
As you proof. Proof de Billy? ))

puumar wrote:
As you proof. Proof de Billy? ))
I gave you a link to the critical temperature.
Got rid of references to the tests,even people who are not in the subject capable of making at least some insights,but apparently it's not your case.
After 6-hour run under load the maximum core temperature did not exceed 86 degrees, which is an excellent result for a graphics card the top-level.

In simple: 30% of maximum rpm, 1500 rpm, 26.6 dBA, GPU 52°C, VRM – 59°C;
Under load: 50% of maximum rpm, 2400 rpm, 39.0 dBA, GPU – 85°C, VRM – 90°C;
In idle mode the fans do not stop.

Idle: 41% of the maximum rpm, 1500 rpm, 31.0 dBA GPU – 33°C, VRM – 40°C;
Under load: 53% of the maximum rpm, 1900 rpm, 39.0 dBA, GPU – 70°C, VRM – 80°C;
In idle mode the fans do not stop.

saa0891 wrote:
I gave you a link to the critical temperature.
And Cho? And here the allowable temperature?
In your proof, even in the pictures show 71 degrees on the chipset after a 6-hour run. Even after this no 80-85градусов, about which you then have to drive it. You merged even their own proof crank.. :S

puumar wrote:
In your proof, even in the pictures show 71 degrees on the chipset after a 6-hour run.

What else can you say,the patient is hopeless.

Simple question, why the map is at a temperature of 65-70 degrees catches the blade ? And the other card for example mining at 80 and above running 24/7 - 1 year or more, equivalent to 3-4 years of normal use in games.
Answer I just mentioned above, the solder which is the hero of the occasion, whether you are 65 or 95 or 100...I will say this, if the developers set for maps of temperature based on the characteristics of the solder temperature would be far beyond 140 degrees, and at 95 is supposedly the maximum that could threaten the map ???

Rasprostranena among the cards problem associated with blade chip, which leads to a black screen, artifacts-possible problem in memory because it uses BGA and solder, but until you look gpu only, a possible problem with the driver installation, more precisely, you will not be able to install the driver, this problem may be caused by the same problem as when artifacts !

What is happening there and how it breaks ! Video card consists of a PCB on which are mounted all the elements, chief of which be the GPU - the graphic processor, since the processor chip it is not possible just to put on a PCB substrate is used for substrate inhabits the chip is the gpu and connects using a BGA layout, that is, housing type surface mount chips, with solder namely lead-free solder, not the best solder, there is a long history of why lead-free, can read =) And why is the blade still occurs at temperatures of 60 and 70 and 80 and above degrees ? And all just have a constant load the same mining, there is a dynamic load that is the game when you play the card temp went up to 80 or 70-65 then you go into inventory and hop load from the card fell to 20-40% and the temperature sharply dropped to 30 degrees and then hop you get out of the inventory and the load is again 100% and esstestvenno sharp heat, and so constantly ! And what about the solders ? He then expanded it to shrink, micro cracks appear in the future, these cracks will lead to the gap that is to lose contact with the PCB substrate, and this Hello Hello artifacts and other pleasures. What's worse he crystal may fall off from the substrate is a black screen not to be treated with warm-UPS need a replacement GPU.

So what I'm getting at, case I remembered a friend took 660 twin froz 2 norms of ahlad more 64-67 she had not seen it cleaned cared but she died over 3 years, with a blade, the artifacts, and how so ? Because the temperature 67 =) That's the lottery that's such a solder, and your songs about 60 degrees on the core is very interesting, especially one in the field of vrm and memory sensors =no)

puumar
Who is lying ?
Spoiler

Mars_City
Yes, I rocked the latter, I connect to the TV, the sound goes through the DVI-HDMI

AnPei
AnPei wrote:
GB seal does not make it not news
Well, that's cool.You can immediately paste to change.And in fact, there were cases when needed to replace pasta, and it was impossible to open due to the loss of warranty

bloomm
And with the dump of the chip card does not recover?

haterwithbeard wrote:
Temperature without 4K 58-62. Stable 60 fps, with occasional dips for a moment to 47.
1660??? What game?
haterwithbeard wrote:
Note the question) whether this Critical temperature for my card? what is the acceptable range?
In the native ha-ke is max temp 82с.
There is an official website of the manufacturer of the video card where you can see all the specs in msi AB, then you have there shows the defaults for overlimit and temperature. As you get closer to the baseline temperature map slowly tratitsya, and if I reach it, trolling will be the maximum. MSI AB allows you to shift the beginning of trolling by sliding the temperature up to maximum, can just the slider to the right until they stop stretching and you will see the maximum allowable temperature in MSI AB put in the base data with the specifications of each chip

Mars_City wrote:
I had a 270 (or rather is), so there was a temperature 88С and everything worked great. Recently it was put for the sake of experiment. Games until 2016 still in full hd drags. Vidyaha topic!
Mars_City wrote:
For the interest will need to swing the firewood and for the gtx 270. I think the map is much that is still can.
Never tried before to write something you think your head? What are you talking about??? At gth 270 was only 896 MB of VRAM and this card was available until approximately 2011, in 2013 it was already unplayable, since it does not have support for dh11 at the hardware level, so no game after 2013 it already does not pull, do not write nonsense

Kyokushin Denis wrote:
Yes, I rocked the latter, I connect to the TV, the sound goes through the DVI-HDMI
So you need a cable hdmi-hdmi or display port, but on the maps of those years, the display port was not

AnPei wrote:
It is sufficient to replace the thermal compound on the adequate and the temperature fell to 10 gr.
There is, producers Chet direct loved gavazova paste to apply. But to change nizya and warranty will fly, and screws a virgin's dirty.

P. Rock
P. rock wrote:
So you need a cable hdmi-hdmi or display port, but on the maps of those years, the display port was not
Do not have.Cable DVI-HDMI works the same on the 7950, as HDMI-HDMI.But on the old XFX4890 somehow the problem is with the sound.He also worked at the old 570GTX until the chip has not fallen off

Kyokushin Denis wrote:
But on the old XFX4890 somehow the problem is with the sound.
I have gth 275 also, there was some kind of trouble, or rather there it was necessary to connect any additional cable from the video thing on the motherboard. In any case, these maps have Neutralny because they don't have support for dh11

Denis Kyokushin
570gtx there might be a big mistake nvidia that cards 8800 /dual-chip/
275 480 760 a good card was hardworking.

Masky76 wrote:
570gtx there might be a big mistake nvidia that cards 8800 /dual-chip/
275 480 760 a good card was hardworking.
You confuse something, probably mean a two-headed gth 590, and 570-ka normal map Pratap was

bloomm
Judging by your logic the processor if so use he, too, must die, but is this not happening

P. Rock
P. rock wrote:
In any case, these maps have Neutralny because they don't have support for dh11
I used it for old games on the old system would have left if not for the sound.

Masky76
Have a 570GTX one disappointing - lacked video memory.It would be at least 2GB.And her pathetic 1280Мб.And this is for DX11
For unknown reasons the old games of 2002 it was better than the 7950.The same Mafia 1 or Return To Castle Wolfenstein.
The first Assassin too, it was better.Pity card.I have a friend bought in 2013м for 5000r.But that's not entirely clear another point:in the last two years noticed a steady job of this card at full speed under load GPU is at 0-1% idle.And it's not mining, it's the driver.Maybe that's why the dead, working in constant tension

Kyokushin Denis wrote:
Well, that's cool.You can immediately paste to change.And in fact, there were cases when needed to replace pasta, and it was impossible to open due to the loss of warranty
It's all personally familiar fillings has long been in few places put. I have one bolt was plastered. Plus there are cushioned bolt hex, and once seen them opened or not.

P. Rock
I'm not confused as these cards have passed through my hands. Exactly 570 from Gigabyte chip fell off after a week.And top-end card from ASUS 8800....did the curve.

Masky76
I will not argue, I 570-d did not own, and 8800 GTS on 320мб I had worked for I think a year and a half, the temperatures that were don't remember, I don't monitor much. In General, a modern iron despite the finer process technology is quite hot. For little videos temperature in the drain mode of operation of coolers over 70 degrees is the norm, but above 80 should not IMHO rise

Kyokushin Denis wrote:
And with the dump of the chip card does not recover?
2 ways.
Heating a temporary solution since the heating of the solder balls expand more, eventually cracks there will only be more it is usually this kind of treatment lasts for 3-4 months, after one warm-up will suffice for less time, and so before the black screen will not come...
Another procedure of REBOL, replacement of chip is done on the soldering station, the technology is similar to landing on the plant. This this procedure is more correct, and the heating method of diagnosis is just not more to make sure Yes it is in the solder and the contact.

P. rock wrote:
Judging by your logic the processor if so use he, too, must die, but is this not happening
Feel the difference the CPU sits on the lga chip graphic on bga..

Apparently did not read carefully, processor base sits on legs of the socket and not solder ! The substrate count of the processor sits on the solders in this problem ! If the count processor sitting on legs how CPU problems would such as a blade would not be, could be the blade of the crystal from the substrate... Which in principle can occur with the CPU. And the processor degrades at high temperatures over time.

bloomm wrote:
processor base sits on legs of the socket and not solder !
And what's the difference? Ali Hosh say that the CPU temperature is 80 degrees? ))