4IFIR vs. Switch OC Suite real benchmark?

[Nakamichi]

No, i don't mix it. The information is placed correctly. You're right, there is no problem with the CPU undervolt (because it just wasn't there) in the old versions, it exists in the new versions (1.7, 1.8) and has not been fixed for a long time. This is a typical situation for 4IFIR, it happens all the time.

I attach a video, no problem.

how did you manage to get ram to OC above 2ghz?

 

[Tomi_desertcat]

sys-clk exposes real frequencies with version 2.0.0. It is just a matter of using 4IFiR loader.kip with original sysmodule (it doesn't affect any voltages and timings) and overlay to see if any of those clocks are faked. Using closed source sys-clk sysmodule (like in 4IFiR) can result in tampered results.

Example what can happen if you don't set correct voltages for CPU (this is not from 4IFiR)
1. Target frequency
2. Real frequency
View attachment 402523

Though the best way is to check them while playing, and this is why I'm working on getting those info from sys-clk sysmodule to status monitor + few other things for 1.0.0 release.

View attachment 402524

so you're working on getting real clock speeds to show up in status monitor+ ?

 

[M7L7NK7]

I guess Modern Vintage Gamer really is not as smart as he appears to be, if he prefers 4ifir.
The difference in performance is massive

He was informed that 4ifir doesn't work, but he kept the video up because gotta get them views

 

[Makc16384]

Yeah, sure. This problem is more than a month old, it's not new. A lot of people have done tests in different games, the result is clearly visible on my videos, the problem exists.

I don’t quite understand what problem we are talking about if 4ifir works correctly for everyone and shows greater efficiency in terms of performance per watt than its competitors?

 

[linuxares]

I don’t quite understand what problem we are talking about if 4ifir works correctly for everyone and shows greater efficiency in terms of performance per watt than its competitors?

The thing is, have it been proven by third party tools? I can claim it's effective but if my software changes parameters to show something else, isn't it just deception then?

This is the problem I have when I read and hear more and more from people that have more know how in the scene than most, I can't help to take their words for it. But also, as long as no one used a frame analyzer. It's not much I can say yes or no about.

That's why the sourcecode could explain a lot, but since he refuse to give it out. It's hard to say what is changed.

 

[meha]

One of these threads again?

Unless someone do an framerate analyze outside the Switch. I won't believe any numbers any longer... Numbers can be faked. Some claim 4ifir do fake numbers, some claim they're real. I don't know what to believe any more.

i don't care if people use 4ifir or oc-suite or whatever. i even say on github that i don't recommend using OCS.

amyways there has been many proofs of 4ifir faking literally everything. so far :

- fake ram clock : 4ifir fakes dram clock so it showed 200Mhz more than actual dram clock. so if it showed 2700, actual value is 2500.
- fake gpu clock : erista has gpu clock limit set to 921, so you cannot go above, but 4ifir had fake erista 998 gpu until i made patch that removed this limit. also mariko has gpu limit of 1305, so you cannot go above unless you do proper patch(which i have done in private version), but 4ifir fakes gpu clocks to like 1.7Ghz
- fake power draw : 4ifir fakes Status Monitor power draw value, so it showed 20% less than actual value, in order to lie about efficiency
- fake dram id: he modified hekate to show every dram id as Samsung MGCL for no reason

- also he posts fake benchmark results

and possibly more

 

[CatcherInTheGrainField]

how did you manage to get ram to OC above 2ghz?

Via the configurator.

I don’t quite understand what problem

I showed it on the video examples. Just look them up.

 

[[Truth]]

i don't care if people use 4ifir or oc-suite or whatever. i even say on github that i don't recommend using OCS.

amyways there has been many proofs of 4ifir faking literally everything. so far :

- fake ram clock : 4ifir fakes dram clock so it showed 200Mhz more than actual dram clock. so if it showed 2700, actual value is 2500.
- fake gpu clock : erista has gpu clock limit set to 921, so you cannot go above, but 4ifir had fake erista 998 gpu until i made patch that removed this limit. also mariko has gpu limit of 1305, so you cannot go above unless you do proper patch(which i have done in private version), but 4ifir fakes gpu clocks to like 1.7Ghz
- fake power draw : 4ifir fakes Status Monitor power draw value, so it showed 20% less than actual value, in order to lie about efficiency
- fake dram id: he modified hekate to show every dram id as Samsung MGCL for no reason

- also he posts fake benchmark results

and possibly more

+he is super toxic.

 

[Nakamichi]

+he is super toxic.

yeah, he is indeed.
Anyone who tries THIS hard to keep his source-code hidden has something to hide.
Still, it helps to finally have the numbers to directly compare.
I suspected that 4ifir was fake for a long time and now we know for sure.
I just hope others find this thread via google and can see the truth for themself.

 

[linuxares]

Wouldn't it be possible to decompile the loader to see what is edited?

 

[Nakamichi]

That would be fascinating.

 

[meha]

i already reverse engineered 4ifir kip file in back old days, which of course showed that it fakes dram clocks
(so back then ST+ had 2666 or higher clocks as default, but reverse engineering it showed that it was in fact 2400Mhz with very tight timings)

also Cooler once released his source files, which of course showed that 4ifir is just fork of OCS with minimal changes that are very dangerous and not good.
(OC Suite is gplv2 project, so anyone can legally reverse engineer 4ifir files. you can go ahead and RE yourself)

all above have proofs lying here and there in discord channels, if you care to look for them.
for example, here is proof that 4ifir 2500Mhz dram is in reality 2300Mhz

 

[CatcherInTheGrainField]

i already reverse engineered 4ifir kip file in back old days

This really happened, but currently 4IFIR does not use fake frequencies. The only technical disadvantage in the latest version that I have identified is an incorrectly working CPU undervolt that loses performance.

Bwt, early 4IFIR followers singled it out for its convenience and ease of installation, just unpack and thats it, however, I made an OC bundle based on the OC Suite, combining all the necessary utilities for overclocking. It is also easy to install and use as 4IFIR, but does not contain its disadvantages.

 

[swutch]

Cooler's claims that he included brick code in 4ifir should be enough to stay away from it.

Closed source with brick code.

 

[Nakamichi]

I just wonder what his motives could possibly be.
It all seems so pointless to me.
Why lie, keep facts hidden, even threaten people with brick-code... over what?
a bad overclocking tool?? stolen code?
for some minor e-fame?

it seems so pointless, doesnt it?

 

[masagrator]

it seems so pointless, doesnt it?

You would be surprised how people that think they have achieved nothing in life can hang on any fame. That's why there are so many sick people doing outrageous stuff on TikTok, etc.

And many of them are getting easily addicted to it, so much that they start believing their own lies that they made up for years. And until they reach lowest bottom of their life, they are not able to discern the truth from imagination - or in better case, they are aware of that, but they believe there is no other way than make up more lies.

That's why discussion with such people is pointless. Either time will come when they will understand that they must change, or they will die believing that they were always right.

 

[BasedKiliK]

I did some deep investigation on Cooler3D and he's a special case. He's actually the one that started an old "Portable Gaming" console Kickstarter that ended up being a massive scam, and even now he's hosting an NFT(!!!) project. Take that information as you will.

Regarding performance across both, I do A LOT of benchmarking, and my own results have determined that 4IFIR is far less stable and has never exceeded performance compared to using the official Switch-OC-Suite setup. In fact, I actually get better performance sticking with the latter (shoutouts to every *actual* dev who's contributed, y'all are fantastic).

Screenshots for proof. Also, look up the PGS Kickstarter project.

 

[Nakamichi]

I did some deep investigation on Cooler3D and he's a special case. He's actually the one that started an old "Portable Gaming" console Kickstarter that ended up being a massive scam, and even now he's hosting an NFT(!!!) project. Take that information as you will.

Regarding performance across both, I do A LOT of benchmarking, and my own results have determined that 4IFIR is far less stable and has never exceeded performance compared to using the official Switch-OC-Suite setup. In fact, I actually get better performance sticking with the latter (shoutouts to every *actual* dev who's contributed, y'all are fantastic).

Screenshots for proof. Also, look up the PGS Kickstarter project.

oh my god! the fake PGS handheld is HIM??
im speechless, i never would have connected the dots there.
i remember when everyone seemingly was talking about that handheld-render and the kickstarter that went nowhere.

and NFTS, too?? Jesus... the typical scammer CV: first crowdfunding then NFT.

 

[Ethelbert]

Damn bros, none of this is good.

I used old 4IFIR 1.5 and 1.6 when I was playing TotK and it seemed pretty good with the graphics plugin mod.
Though eventually it started being pretty unstable as I was trying the new versions, so I settled back to 1.5 and finished the game with it.

Since then I of course updated everything and switched to OC Suite.

 

[Chrisssj2]

You're all are nuts. I get amazing FPS with 4ifir.