Can anyone weigh in on what they think would be "good" clocks for an FTW3 hybrid, running at "fine" temps of 40-50?

I'll see if I can find the ones with LN2 and XOC.
It went faster with those
It was an excellent card , still is
Pretty sure I got it up to 2200

Im work chat later

Jiberish001
If 2100 is being considered "good" clocks, with "perfectly fine" temps of 40-50, then what would you consider to be bad clocks at those same temps? How about more like 42c.... What do you think would be a bad clock in need of inspecting for problems?

I WISH I got close to the clocks you two are talking about. I would have previously thought they were awesome when combined with the temps I get, before I learned that apparently what I used to think was awesome temps is now just fine.

Anything less than the advertised clocks would be bad.

Sajin
Anything less than the advertised clocks would be bad.

I'm asking for personal opinions. Not PR.
OF COURSE anything less than advertised clocks would be bad, but I want to know what YOU would consider less than good. That answer sounds like a copout answer I'd get from a PR shill.

My card is advertised as "1755MHz Boost Clock."

You telling me that if YOU got 1755 as your max boost that you'd be okay with that, and only if it were lower would you say it was bad?

As an addendum (sorry, but if I edit my post it gets removed for some reason), if 2100 is "good", then that's a big leap from 1755 advertised to a "good" 2100.

That's much better. Thank you. :)

Now, if your clock was running in the bad zone, even though still in the advertised zone, you would look into any potential problems, right? Can I ask you what you would look for next that might be the problem and would be something you'd considered to NOT be acceptable from the vender?

spit051261
I would put it down to silicon lottery .

Makes sense, if you knew it was all working properly, sure. But wouldn't you check to make sure it WAS working properly? Surely you wouldn't just accept such bad luck right from the start, would you?

Jiberish001
Now, if your clock was running in the bad zone, even though still in the advertised zone, you would look into any potential problems, right? Can I ask you what you would look for next that might be the problem and would be something you'd considered to NOT be acceptable from the vender?

I would only test the gpu to make sure everything is working without artifacts. A low clock doesn't mean your card has a problem. It only means you didn't get a chip that clocks as high as others.

Of course it doesn't mean there IS a problem, but a problem can cause it to run slower. I'm not asking if you would think there is a problem. I'm asking if you would look for any POSSIBLE problem if it was running slower than others, and what you might potentially find that you would consider an issue that the vender should be held accountable for.

Please let it be clear that I am NOT talking about my card during any of this.

As an aside; do you think it is strange that evga lists the power requirements for their cards, but they don't list the power draw that particular card should draw?

Ok
Don't know what to suggest.
I do but I always get stick when I talk about it

Silicon lottery, and temps out of control, I can understand. But when you say "bumping into the power limit" what exactly are you talking about?

Are you putting blame on the user not having the power to give the card, or blaming the card for being limited on how much it can draw regardless of how much the user has to offer?

Also, as a point of curiosity,... I have a 1000 watt psu, and EVGA states that the power requirement for my card is a 650 watt psu with two available power dongles. (yes, I have two attached) Isn't that really low for such a card as this? And why does it say under that "Total Power Draw : 250 Watts" ???  That's definitely not the card. The card has a target of 300, with a max of 373. What is this 250 nonsense?

I just realized that what evga is calling "total power draw" is just the card's TDP. Honestly, I didn't think that graphic cards dealt with the whole TDP stupidity. I thought that was just a cpu thing.

OH MY GOD. Precision. :(

So, I was messing around with the fan curves of precision, and noticed something weird whenever I tried to increase my fan curves. I have (had) two fans on my radiator, hooked up to a Y split and then into the card. I was trying to do some tests and I wanted to set my radiator fans up to 100%, but increasing the curve in precision lead to something weird happening. Whenever I set the rad fan(s) to 100%, (first it took FOREVER to initiate),  it seemed as though it only wanted to work for one of the two fans connected to the Y split. It kept revving, up and down, up and down. Never reaching full speed as if the other fan was holding it back or something.

So first I did an experiment. I connected a fan to the card's extra fan plugin (just sitting there off to the side of my case) and loaded things up and sure enough it was now showing "fan 1, fan 2, fan 3". (vrm, radiator, new fan)  But then here is where it gets weird.... I was trying to figure out which fan was which, so I was messing around with the speeds. First I changed the speed of "fan 3", thinking that this must be the new fan I added....NOPE. It was acting like it was the two fans I had on the radiator with the Y split. BUT, it was acting weird again, as though it only wanted to increase the speed of ONE of those two fans. Revving up and down constantly as if one of them was holding the other back. So then I tried increasing the speed of fan number 1, thinking maybe it added the new fan in front of the others. AT FIRST it looked as though "fan 1" was the vrm fan, because as I changed the speed I could hear the very distinct sound of the higher pitched vrm pan revving up. so THEN I was like "okay, this is weird, lets start over"..... So I hit the "default" button to reset everything and got up from my chair to put my ear to my (open) case. Then I turned up "fan 1" again to confirmed it was the vrm fan, BUT this time it started revving up that extra fan I had off to the side, as if it had changed what it now considers to be fan 1, 2, and 3. It was like the program was constantly scanning the fans (why it was taking so long) and then deciding which fan was which.

So I powered everything down completely (full cycle with power unplugged and any residual gone) and unplugged the two rad fans from the card and put them onto my motherboard. I dialed them up to 72% in my bios, which equaled out to about 1400+ rpm on both (can probably go higher without too much sound annoyance), and then rebooted everything......

I ran a scan in precision and then was able to get a score of 104, when before I could NEVER get anything above a 99.

I total get how important temperatures are with these cards, but I would have thought that IF the card needed the fans to spin faster when it was getting too hot, that it would do so, but apparently it does not do this anywhere near soon enough, and also cannot even properly see what the fans are doing anyways.

Before, when I had the two fans hooked to the card, I would get temps a little over 50c (like maybe 52 at absolute max) whenever I'd do a scan. But this time (with the motherboard controlling fans to keep them running at 1400+ constantly) I never saw temps above 48c.

More testing is needed, but I am now VERY disappointed with this precision software.