xFluing

9 hours ago, -rascal- said:

 

VSR works with 1680x1050.

Best to use 16:10 resolutions...so like you said, like 1920x1200 and 2560x1600.

2560x1600 is pretty heavy on the GPU though, don't think a R9-270X can handle that too well...

Thanks a lot! I was mainly shooting for 1200p, but I'm sure R9 270x will be more than enough for 1600p on older games, besides I don't mind dropping the settings to medium, which is gonna help the card a lot in higher resolutions

1 minute ago, -rascal- said:

 

From what I can see, VSR is supported for 1680x1050.

I don't think the VSR page is the most up-to-date.

 

Regardless, I can double check in a little bit -- I have a spare 1680x1050 monitor.

Thanks, that would be great!

2 minutes ago, -rascal- said:

 

Yes.

What VSR does is render the images/frames at...1920x1200, but squashes it down to...in your case... 1680x1050.

Your GPU is still putting in work as if it was pumping out frames at 1920x1200 resolution.

 

Your image quality will be better, but your GPU just has to work harder.

What I'm getting from the AMD site is that you can use VSR if you have a 1920 x 1200 monitor, and I don't, and I want to know if I can use VSR with a 1680x1050 monitor even if it's not mentioned

I plan to upgrade my card to something more modern, like R9 270x and the AMD site states that they only support certain resolutions for use with VSR, and one of those is 1920 x 1200, which is the next step up from my 1680 x 1050 monitor, and I was wondering if I could run VSR with a 1680 x 1050 monitor

1 hour ago, Mick Naughty said:

So idle it down clocks like it should? Could change a setting like Nvidia has for multi monitor and performance modes. 

The memory does not downclock ever as far as I'm aware, so no, it doesn't downclock as it should, because it's not the core clock, it's the memory clock

I have a Radeon R7 250, and ever since a power surge I've been having this odd issue where my memory clock on idle will downclock itself whenever I mess around with even the slightest option (+/- 1MHz core / memory clock) which causes my secondary monitor to flicker like crazy.

Under load it's ok, it runs at the full clock, but it's when it's idle that it's causing the issues.

It's not the drivers, I've had it through 4-5 versions now.

On 2/21/2018 at 8:55 PM, ryanf257 said:

Have you tried uninstalling AMD settings apps?

 

I didn't, because I never thought of this as an issue, I did re-install drivers recently but I didn't bother to check because this behaviour seems completely normal.

Same thing is happening to me, it's not as bad, but I'm also getting "3D mode" clocks and about 20-30% usage when messing around in radeon settings.

Same thing is happening to me with my R7 250, I wouldn't worry about it. There is no reason for the fans to spin up max during boot yes, but it's not bothersome in any way, you're not playing games or anything during boot so it shouldn't matter.

15 minutes ago, Else89 said:

Ok, so i've decided to take the 1060 and partial refund. Having a working system is more important to me than a few frames. Yes i'm going to drop down in settings but it's worth it to actually play games again. 

 

The partial refund I am getting will probably be put towards a new card when the new stuff drops. just disappointed that I can't get my 1070 back.

Dropping in settings is not a big deal anyway, in modern games most of the time you won't even notice the difference unless it's side-by-side comparison, I found that for me Medium is the sweet spot.

I have an r7 250, which I successfully overclocked to 1150 core clock and 1250(?) memory clock, all was stable and good until a power outage, after which I began seeing weird flickering on my secondary monitor, and I thought the whole video card was fucked, however I found out that the power surge is what caused the overclocks to do that to the monitor. The secondary monitor is plugged in the VGA port, and the main one in the HDMI port, the main one having no issues like that whatsoever. The weirder thing is that it does this flickering even when underclocking, basically whenever i mess with the default values whether it be increasing or decreasing.

I've heard this could be caused by fucked VRMs on the graphics card, but I'm not sure really. It works fine with +20% power in furmark (+20% in furmark is the only increase in power that allows the gpu to fully ramp up to the base 1050 speed, otherwise it power throttles down to 1000, and it only happens in furmark). There is no artifacting going on anywhere while the flicker is happening on the secondary monitor, so what can I do to fix this? A workaround that I thought of would be to load an OC profile before playing a game then just turning off the monitor and deal with it that way.

Yes, last time I checked (which was before the power outage) these overclocks WERE stable

Regardless I would like to know exactly what happened. Fucked VRMs or is someting else fucked? Maybe just the VGA port itself? I can't test the DVI port, I don't have any DVI-D monitors or any DVI-D to VGA adapters for that matter, I only have DVI-I ones.

Looking back at the video, I also noticed that messing around with the clocks (both core and memory) for some reason it underclocks the memory to 150.

Here's what it looks like on video

30 minutes ago, Narnash said:

Don't try it with casted ones, you will most likely destroy the fan. Most modern fans today come cast or closed, they sould be lubed up for their full life time. 

 

If your fans feel rough or make wierd noises than replace them completely or send them in if you still have warranty.

It's starting to make rattle noises, but only starting at 25% fan speed, then taking a break until 40-45%, even though I'll never hear it since it kinda never reaches that high RPM anymore ever since I changed the thermal paste on it, I still want the rattle gone.

I want to lube the fan of my card's cooler, but the cap on the fan is cast, meaning I can't just pop it off to reach the bearing and oil it up, what can I do to prevent locking up the fan with molten plastic? I want to use the solder iron to melt it off and put some oil in this bitch.

1 minute ago, VikingGaming said:

Might be bad caps on the GPU. That can happen over time.

I've checked, nothing is swollen.

I have this old Nvidia Quadro FX 3450 card, and a year or two ago, albeit in its original Dell workstation configuration, was fine, it wasn't showing up with anything.

However, I recently got a power supply that supports pci-e connectors and decided to test it out in my system with a relatively more modern CPU, miles better than the dual-core Intel the workstation had (I believe it was an E6300), as I believed an Athlon x4 860k would remove any CPU bottleneck issues.

The problem here is, even though it sat untouched for a year or two, the GPU is artifacting now (windows did manage to install a generic driver for my current amd gpu before shut down, so it might be that), but I'm getting artifacts even before the system starts up which makes me believe it's a hardware issue, now I have no idea if it's caused by the GPU dying, or because i plugged it in a non-OEM motherboard (can artifacting even be caused by that?)

I tried both DVI connectors, and for both I got the same results (the first time I turned it on the artifacting was way worse, with purple dots all over the screen, but it seems to have stabilised to what it looks like in the pic)

Regardless, reflowing isn't an option unless I get a heat gun which I haven't looked into just yet, just wondering what could be causing this, and how it could have broken down just by sitting untouched.

1 hour ago, Crunchy Dragon said:

Could be, normal load temps are around 60-70 so don't be alarmed if it does rise to that temperature. As far as voltage goes, you should be fine.

OK tested stuff out and surprisingly I got 51-52 degrees in gta 4 (which is odd, I always got higher temps in gta 4 than aida), no extra voltage, didn't go past 1.46, though it does seem to idle at 1.47 but should still be fine.

1 hour ago, Crunchy Dragon said:

Yeah, for a long time nothing Intel had to offer reached that high up until the 6th Gen Extreme Edition Core i7s, if I recall correctly.

 

Good luck on your overclock!

OK, I'm up to 4.2 GHz now, seems stable, still running a stress test and these are the numbers I'm getting.

That 1.456 voltage is just a tiny spike every now and then, now the question is will a game like GTA 4 cause the voltage to rise even more, seeing as I get way higher temps running GTA 4 than an AIDA stress test.

1 minute ago, Crunchy Dragon said:

Yeah, for a long time nothing Intel had to offer reached that high up until the 6th Gen Extreme Edition Core i7s, if I recall correctly.

 

Good luck on your overclock!

Thanks!

2 minutes ago, Crunchy Dragon said:

I wouldn't expect that to happen, another thing to keep in mind is that FX series chips like the one Jay had consume a pretty much the most power out of any CPU released in the years when AMD still manufactured them so it's less likely that your 860K would reach that point, having a much lower power consumption.

True, I've seen FX chips having a TDP of up to 125-220w, besides the fact that they're nicknamed "chipset fryers", a friend has an FX8000-series, and his chipset runs so fucking hot it's not even funny.

Just now, Crunchy Dragon said:

You should be pretty good to go, it is after all only a 300Mhz boost from the base clock to hit 4Ghz. I boosted my A6 by just over twice that(800Mhz) to reach 4.4Ghz.

 

As long as your overclock is stable and you're seeing good temperatures, I wouldn't worry about it.

Yeah but I'm still worried that even at 4.2 I might hit a threshold where the voltage would keep rising and a restart would fry the cpu, like it happened to jay (however I don't think that should be a problem since I can avoid that by resetting the CMOS)

Just now, Crunchy Dragon said:

I wouldn't worry about it, usually auto voltage doesn't let you boot if you set a clock speed that requires higher voltage(in my experience). Most people recommend you don't go above 1.5v as a rule of thumb.

Yeah, 1.5 is below what I'm trying to stay.

3 minutes ago, Jurrunio said:

Max temp to max voltage:

85C: 1.3V

80C: 1.35V

70c: 1.4V

60C: 1.45V

 

Dont go any higher, that's dry ice territory

 

In terms of temps, I've always wondered what do I look for? Overall temp or core temps? I never got the difference.

6 minutes ago, Crunchy Dragon said:

There's nothing wrong with overclocking, even if you're not a professional that could use the time saved. I was more or less forced to overclock my A6-5400K just to help it keep up with Windows tasks and other basic workloads since I couldn't afford a new CPU at the time. And believe me, it helped. Overclocking and seeing better results isn't limited to professionals that could use the extra time.

 

That's a lie, the only way to actually shorten the life of a CPU from overclocking is to run it at higher than supported voltage and/or higher temperatures.

 

@xFluing: JayzTwoCents has quite a few overclocking videos you can refer to, such as this one which lets you know exactly what you're getting into when you overclock.

 

In terms of the question in your original post "would going past 4Ghz fry it at auto voltage?", I pushed an A6-5400K(3.6Ghz base) to 4.4Ghz with the stock cooler and that's running fine to this day. I didn't set any voltage, just left it at auto. If your CPU has adequate cooling, I wouldn't worry about it at all.

Also yeah, it has more than adequate cooling, great air flow with a Gammaxx 400 cooler on top.

Just now, xFluing said:

Yes, thanks, I know about overclocking, I've watched those videos, thing is I don't know how far I'm gonna push the voltage on the CPU if I go past 40 in the cpu multiplier, because it automatically adjusts the voltage and I have no control over it and I don't want to fry it by going into the 1.6V territory (something that did happen to Jay when he was overclocking his FX I think, which ended up in frying it)

OK nevermind, I did not see that last part because of the video.

2 minutes ago, Crunchy Dragon said:

There's nothing wrong with overclocking, even if you're not a professional that could use the time saved. I was more or less forced to overclock my A6-5400K just to help it keep up with Windows tasks and other basic workloads since I couldn't afford a new CPU at the time. And believe me, it helped. Overclocking and seeing better results isn't limited to professionals that could use the extra time.

 

That's a lie, the only way to actually shorten the life of a CPU from overclocking is to run it at higher than supported voltage and/or higher temperatures.

 

@xFluing: JayzTwoCents has quite a few overclocking videos you can refer to, such as this one which lets you know exactly what you're getting into when you overclock.

 

In terms of the question in your original post "would going past 4Ghz fry it at auto voltage?", I pushed an A6-5400K(3.6Ghz base) to 4.4Ghz with the stock cooler and that's running fine to this day. I didn't set any voltage, just left it at auto. If your CPU has adequate cooling, I wouldn't worry about it at all.

Yes, thanks, I know about overclocking, I've watched those videos, thing is I don't know how far I'm gonna push the voltage on the CPU if I go past 40 in the cpu multiplier, because it automatically adjusts the voltage and I have no control over it and I don't want to fry it by going into the 1.6V territory (something that did happen to Jay when he was overclocking his FX I think, which ended up in frying it)