buildzoid

Doesn't Asus have an external BCLK generator on the ROG boards? Would it not be possible to force the chip into getting it's BCLK from the external generator?

So do we have 2 people who would be able to stream on the night of 23rd? I'll be doing the casting.

I am very smart /s I soldered the completely wrong pins no wonder it still won't start.

290X VRM is really really hard to circumvent because the IR3567B is full of safety features. Also the stock R9 290X VRM isn't all that bad. At 100C MOSFET temperature you can push 528W to the core. On LN2 when the VRM is running colder you can push 720W to the core. The only reason why you would want to put an E-power on an R9 290X is that the stock VRM is pretty noisy. Also you can always just buy a 290X Lightning and then you don't need an E-power at all.

I've managed to hold an audience for 3-4 hours with a good session. The commentator in this is way way way more important than say in football because in football people roughly understand what it is that they are watching. With this the commentator will have to do a whole lot of explaining of various topics to keep them interested. Of course the even will be split into 2 parts. Part 1 would be preparations and then Part 2 would be actual benching so people who don't want to watch the prep can just wait until part 2 is on and watch that.

The idea behind this is that it's going to be short say 3/2 hours during the 3/2 hours you do your setup and benching(you agree on cooling before hand). The job of the commentator during this time is to keep the audience interested. This means that they will talk about the people doing the benching, the hardware being used and the of course the scores and techniques being used by the benchers. To get interesting info on to talk about during the stream pre stream interviews with the participants might be necessary but these will not be public and there won't be any interviews after the stream either. If you don't want to get interviewed you just gotta send the commentator some info about yourself so that they have intresting things to say. So things like when why and where you got into extreme OC. Favorite system. The amount of binning you do. Past achievements. The kind of thing that people watching you are interested in. Your job as the bencher is to get highscores and have fun. The commentator needs to hold the audience. @Xyala would it be possible for us to discuss technicalities of streaming and commentary over skype or something?

subbed

The idea behind this is that people can compete with each other over long distances. So hypothetically I can compete from my basement in the CZ against someone in the US.

OK so I've been doing overclocking live streams for a bit with varying levels of success. On one of the live streams I had the idea that it would be much cooler if there was 2 people competing against each other. So that's what this thread is about. If you want to live stream overclocking you optimally need 2 cameras one to watch the system and one to watch your screen. If you're benching 2D you can use a capture card however if you setup a camera well you do not run into any unusual issues that you might get with a capture card. Ideally a stream like this would need 3 people to actually work. 2 people benching and 1 person doing commentary and maintaining a score/settings overlay on the stream. Whoever is doing the commentary will need quite a bit of download speed since they will be getting 2 external streams that they need to combine into one on their side and upload that to twitch or whatever. Right now I only have access to my laptop. I'm willing to do commentary+stream mixing for any 2 people who can and want to do the benching. We'll first have to do a trial run to see what the best stream setup is as far as the mixing and overlay goes. Then if that works out we can arrange a time and the 2 people benching agree on the benchmarks and systems and we do it for real.

what a set up from FatBoyNotSoSlim and perfect finisher from Massman!

Did you try running asymmetrical clocks? As in each gpu running different core and mem clocks?

I assume this is stock voltage right? I made a Vmod guide for the reference PCB Fury and FuryX: http://cxzoid.blogspot.cz/2015/09/r9-fury-tri-x-and-fury-x-vmod-guide.html . HBM on air doesn't seem to scale but the core clock goes well with voltage. I was able to bench as high as 1180mhz on 1.3V.

That actually made me LOL

Mercury wouldn't work since it has stupidly high surface tension it will not wet anything and so the thermal conductivity will be crap.

This chip is so so so very bad

I wish I had bigger cooling towers on my TECs I could probably push a couple MHz more also gotta delid the CPU.

How's the VRM on the FTW also what controller does it use? I think I might get one for starting LN2 on GPU or I can throw it on the chiller.

Both the Fury and Fury X are listed as having a stock clock of 1100mhz. The actual clocks are 1050 for the Fury X and 1000 for the Fury.

Do you have a full datasheet for the IR 3567B? I've been struggling with making everything work with just the public one and the IR 3567A datasheet(seems to be basically the same controller and the datasheet is a little more detailed)

Thanks for pointing that out. The OCP is only raised by 10% so it shouldn't be too much of an issue. Either way I'm going to be making a new heatsink for the VRM since the first one I made only kept it at around 80C on 1.35V with a 4K RPM fan on it.

So I decided that now that I have a Fury I could practice my volt modding on my 290X and this is the end result: It still works but it no longer fits the stock cooler so I can't test if the volt mod actually works until I get an alternative cooler for it. Either way I'm glad I did the 290X first because now I won't make the same mistake when I do my Fury. Can anyone guess what I did?

Can it do 5Ghz on about 1.5V with water cooling?

I've asked the guy for a CPU-z shot. IDK if he will deliver.

I heard someone has a 5.2Ghz 7870K.

Yeah I just returned a chip that needed 1.45V to boot 4.6Ghz. I'm really really unlucky with most of my CPUs(I had a 750K that needed 1.75V for 5Ghz and a 370K that needed 1.65V for 5Ghz). Hopefully my next chip can hit 5GHz on something bellow 1.6V.