canceled​

 

[cs2020]

You’d need lab equipment for that.

I see... so it's not so easy as it looks.
btw, the overlock works itself, maybe the improvement % isn't exact, but on 4ifir in kakarot i get solid 60 FPS with maxed graphics by cheats, and on os suite I have also improvements, aroud 50-54 fps with same setting, so I can't understand why there is all the hate against Cooler, he also took the development of oc suite, because ctcraer caused Kazu to quit because of his bulling...
Cooler didn't bullied anyone, or not?

Just posting facts, don't take it personally (to every1)

 

[Lumince]

Took the development of oc suite, because ctcraer caused Kazu to quit because of his bulling...

Don't claim rumors as facts. CTC has already stated that it wasn't him.

 

[Lamcza]

I see... so it's not so easy as it looks.
btw, the overlock works itself, maybe the improvement % isn't exact, but on 4ifir in kakarot i get solid 60 FPS with maxed graphics by cheats, and on os suite I have also improvements, aroud 50-54 fps with same setting, so I can't understand why there is all the hate against Cooler, he also took the development of oc suite, because ctcraer caused Kazu to quit because of his bulling...
Cooler didn't bullied anyone, or not?

Just posting facts, don't take it personally (to every1)

Looks like a solid improvment 60fps stable from 50-54 nice

Don't claim rumors as facts. CTC has already stated that it wasn't him.

yea we believe you totally for 100% we believe you lol i can't even...
We saw from the discord screenshots how trustworthy person you are and what you think about ppl on this forum and this forum so pls end it and let's keep this thread clean of offtops and dramas ; )

 

[Lumince]

Looks like a solid improvment 60fps stable from 50-54 nice



yea we believe you totally for 100% we believe you lol i can't even...
We saw from the discord screenshots how trustworthy person you are and what you think about ppl on this forum and this forum so pls end it and let's keep this thread clean of offtops and dramas ; )

My opinion on this forum gets worse and worse the more yall reply. Good luck and I hope your switch doesnt die on you.

 

[alex61194]

im bored of all this drama, release it or not but stop the drama pls

 

[cs2020]

Good luck and I hope your switch doesnt die on you.

bro, why so toxic?

 

[Foxi4]

I see... so it's not so easy as it looks.
btw, the overlock works itself, maybe the improvement % isn't exact, but on 4ifir in kakarot i get solid 60 FPS with maxed graphics by cheats, and on os suite I have also improvements, aroud 50-54 fps with same setting, so I can't understand why there is all the hate against Cooler, he also took the development of oc suite, because ctcraer caused Kazu to quit because of his bulling...
Cooler didn't bullied anyone, or not?

Just posting facts, don't take it personally (to every1)

300% is just an obscene figure - that’d be going from 30 to 90FPS after the overclock, which is silly to even think about. Fact of the matter is that the reading is erroneous or the math is fucked up. That *doesn’t* mean that there’s no increase in performance, it’s just not that high.

 

[alex61194]

300% is just an obscene figure - that’d be going from 30 to 90FPS after the overclock, which is just silly to even think about. Fact of the matter is that the reading is erroneous or the math is fucked up. That *doesn’t* mean that there’s no increase in performance, it’s just not that high.

marketing?

 

[Lumince]

bro, why so toxic?

I see no toxicity with the statement you quoted.

 

[Foxi4]

Let’s stay on-topic and not make the thread an indictment against anyone posting in it. You guys have a clean slate, so keep it civil. If anyone’s under the mistaken impression that I was joking around in my earlier post, further personal bickering is a quick way to test that theory. If the post isn’t about 4FIR, it doesn’t need to be posted here.

 

[Xandroz]

Sorry to ask but which config can i use safe with the oled varient on atmosphere latest

 

[Venomzver]

300% is just an obscene figure - that’d be going from 30 to 90FPS after the overclock, which is silly to even think about. Fact of the matter is

friend you are partly right, but everywhere they say so. open the page with gpu or ram overclocking. here everywhere they talk about the performance of gpu. if earlier it was at 430 megahertz, now it is 810 megahertz. it is clear that productivity will not increase twice

Post automatically merged: Mar 17, 2023

it's like the engine of a car, if it is 300% more powerful, this does not mean that the car will go 600 km per hour instead of 200 km per hour. I hope I explained correctly

 

[Foxi4]

friend you are partly right, but everywhere they say so. open the page with gpu or ram overclocking. here everywhere they talk about the performance of gpu. if earlier it was at 430 megahertz, now it is 810 megahertz. it is clear that productivity will not increase twice

Post automatically merged: Mar 17, 2023

it's like the engine of a car, if it is 300% more powerful, this does not mean that the car will go 600 km per hour instead of 200 km per hour. I hope I explained correctly

Then it’s not a 300% performance increase. MHz is a measurement of frequency, not a measurement of performance, so you’re undercutting your own argument. Wow, you’re making me feel old - I feel like I’m back in the MHz Myth days. Lemme think, how do I ELI5 this…

Here’s the thing. This is how it actually works on a curve. Imagine you have a theoretical benchmark and a theoretical processor running at 100MHz, for ease of calculation, and in said benchmark you get a score of 1 - that’s your baseline (stock) figure. You overclock that CPU by 50MHz (bold move, Cotton, but I have seen that on ye olde CPU’s back when 100MHz was considered “a lot”) and now you’re at 150MHz, that’s a 50% overclock. You run that benchmark again and, as it turns out, you get a score of 1.25. Let’s put those figures together:

100MHz = 1
150MHz = 1.25

You overclocked the chip by 50% (100MHz -> 150MHz) and your performance (the amount of work done in the same timeframe) increased by 25% (1 -> 1.25). You’ll notice that that’s not perfect scaling, you’re not getting 50% extra performance from a 50% higher frequency - scaling is logarithmic, so you get less and less performance the further you go up the curve. The MHz figure doesn’t matter - that’s not performance, that’s just how many cycles a CPU does in a second. What matters is the amount of work done in a fixed timeframe, usually per second (MIPS, FLOPS etc.) - makes sense?

Now, collect data points like this until you feel sick to your stomach and eventually you’ll chart a more realistic performance curve. Your performance increase is the percentage difference between performance at stock and performance at your desired clock. Saying that your performance increased by 300% because you’re running the core at 3x stock frequency (which is not to be confused with base frequency of the chip - the Switch runs underclocked by default, Nintendo chose to do so to save battery life) is just not true - your performance increased by however much your score increased.

To put it in your engine analogy, your engine has stock horsepower. You can do a bunch of shenanigans to the engine and you might end up with more horsepower. Your increase in performance will be the difference between how much horsepower you had at the beginning and how much you have after you modified it, or fed it special sauce, or whatever it took to get more power out of the same engine. Nobody cares how many RPM’s the engine is spinning at if it doesn’t translate to meaningful power.

 

[Cooler3D]

Then it’s not a 300% performance increase.

Frankly, I'm confused.

You have come to the conclusion that I lied when I described 4IFIR. Which also means that from your point of view, I am misleading the forum participants (by declaring obviously impossible characteristics). But instead of banning me, you justify me and stand up for my projects...

Don't get me wrong, I am very grateful to you, and deeply appreciate your gesture of support... Just... I sincerely believe in the validity of my statements. I understand how implausible such a statement sounds, but if you don’t mind, I will not only prove that I was not cunning, I will also gladly explain in detail how I managed to achieve the declared characteristics. If you are interested in this in principle, and I do not take your time for meaningless things.

Can I?

 

[Foxi4]

Frankly, I'm confused.

You have come to the conclusion that I lied when I described 4IFIR. Which also means that from your point of view, I am misleading the forum participants (by declaring obviously impossible characteristics). But instead of banning me, you justify me and stand up for my projects...

Don't get me wrong, I am very grateful to you, and deeply appreciate your gesture of support... Just... I sincerely believe in the validity of my statements. I understand how implausible such a statement sounds, but if you don’t mind, I will not only prove that I was not cunning, I will also gladly explain in detail how I managed to achieve the declared characteristics. If you are interested in this in principle, and I do not take your time for meaningless things.

Can I?

I didn’t say anything about lying. I said that you (almost definitely) messed up the math on that one, or you’re treating an increase in frequency as an increase of performance, or something else along those lines. I think you should be allowed to share your tool and if people like it then all the power to them. I don’t think your performance claims are real though - I think they may be a result of measurement error or bad math, or just a misunderstanding of how to calculate an increase in performance. Deception requires intentional malice, and I have no evidence of that.

EDIT: Let me give you an example. Let’s say you run a game with an unlocked framerate and fixed resolution at stock values and you take measurements at certain points - those are your data points. You’ll have your peak (highest number), bottom (the absolute lowest number, usually described as 0.1% lows) and an average of all of the points. Let’s say you average 45 FPS. Then do your overclock, and measure at the exact same points. Compare those two figures. Let’s say you’ve gained 15 extra frames on average, (that’s very generous by the way) and can now run at 60FPS most of the time. That’s a 33.3% increase. That’s it. The more points you put on a graph the higher the granularity of your analysis (most people just go with 1 second = 1 datapoint since we’re talking about frames per second here). This is assuming the figures reported by the system are true, which they may or may not be. That’s how you’d measure the impact of your app fairly.

100% = 45 FPS
X% = 60 FPS
100/45*60 = 133.3%

You’re running the game 33.3% faster than before, that’s your performance gained. To get a 300% gain you’d have to run the same section at 135 FPS, which I’m sure you’ll agree is not realistic (or even possible, for that matter).

 

[Cooler3D]

I didn’t say anything about lying. I said that you (almost definitely) messed up the math on that one, or you’re treating an increase in frequency as an increase of performance, or something else along those lines. I think you should be allowed to share your tool and if people like it then all the power to them. I don’t think your performance claims are real though - I think they may be a result of measurement error or bad math, or just a misunderstanding of how to calculate an increase in performance. Deception requires intentional malice, and I have no evidence of that.

If you don't mind, I will describe my testing methodology, supported by comparative examples that anyone can reproduce in their own. If I made a mistake in the calculations, I would like to find out for sure, I do not want to mislead anyone with incorrect statements. However, if you are not at all interested in how I came to this amazing conclusion, and what I was wrong about, I will not waste your and my own time on this. Moreover, there are enough tasks without it.

In any case, thanks again.

 

[Foxi4]

If you don't mind, I will describe my testing methodology, supported by comparative examples that anyone can reproduce in their own. If I made a mistake in the calculations, I would like to find out for sure, I do not want to mislead anyone with incorrect statements. However, if you are not at all interested in how I came to this amazing conclusion, and what I was wrong about, I will not waste your and my own time on this. Moreover, there are enough tasks without it.

In any case, thanks again.

I just gave you the correct methodology - in order to get a correct measurement (or as correct as can be), you’d have to use a controlled environment (fixed resolution, fixed workload, like a demo animation, so it can be replicated relatively accurately. In the absence of that, a similar enough path on the same level), calculate the average pre and post modification and then use that difference in total frames as a percentage difference. I even gave you the (broad strokes) math.

The reason why I mention 0.1% lows is because that’s what the human brain perceives most readily. If the game runs relatively smoothly but in *some* sections it randomly starts dropping large numbers of frames, it will have the appearance of running choppy. This is why a lot of games lock the framerate with a cap in the first place. The developers know that if the game will, at worst, run at 30 FPS when the screen gets busy, they’ll cap it at that because it “feels” smoother than randomly jumping between 60 and 30 - that’s jarring. That’s a measurement of comfort though, so an average is good enough to demonstrate.

Hope that clarifies some things, and why people raise an eyebrow at your claims. Maybe it’s a language barrier thing, I don’t know, but I’ve been overclocking since the days of MS-DOS and a 300% increase in performance is pie in the sky. Again, not calling you a liar, but you are doing something wrong, or phrasing it poorly.

 

[Cooler3D]

I just gave you the correct methodology - in order to get a correct measurement (or as correct as can be), you’d have to use a controlled environment (fixed resolution, fixed workload, like a demo animation, so it can be replicated relatively accurately. In the absence of that, a similar enough path on the same level), calculate the average pre and post modification and then use that difference in total frames as a percentage difference. I even gave you the (broad strokes) math.

The reason why I mention 0.1% lows is because that’s what the human brain perceives most readily. If the game runs relatively smoothly but in *some* sections it randomly starts dropping large numbers of frames, it will have the appearance of running choppy. This is why a lot of games lock the framerate with a cap in the first place. The developers know that if the game will, at worst, run at 30 FPS when the screen gets busy, they’ll cap it at that because it “feels” smoother than randomly jumping between 60 and 30 - that’s jarring. That’s a measurement of comfort though, so an average is good enough to demonstrate.

Hope that clarifies some things, and why people raise an eyebrow at your claims. Maybe it’s a language barrier thing, I don’t know, but I’ve been overclocking since the days of MS-DOS and a 300% increase in performance is pie in the sky. Again, not calling you a liar, but you are doing something wrong, or phrasing it poorly.

Thanks for the tips, I'll use them. If I get something worthy as a result, I will invite you to evaluate.

... and if it doesn't work out, then I'll change my nickname to CoolerX5

 

[Chrisssj2]

Why is exosphere.ini so different then rentry switchhackingiseasy, and no hosts.txt in atmosphere?
Im odc about no ban my switch and dont trust that xD

 

[Cooler3D]

Im odc about no ban my switch and dont trust that xD

Zero known cases of console ban on 4IFIR.
If you are calmer with hosts, you can safely prescribe. It shouldn't get worse.