Status
Not open for further replies.

smf

Well-Known Member
Member
Joined
Feb 23, 2009
Messages
6,643
Trophies
2
XP
5,861
Country
United Kingdom
This all doesn't belong in this thread anymore. Where did Foxi4 say the Switch draws Voltage? Voltage is supplied by the VRM's and it draws the current. It can, however, ask the VRM to supply a certain amount of Voltage. This is usually called the VID. That's not the point here, though, and neither is it mentioned anywhere.
I'm not sure I want to go through all his comments... but this is one place.

Of course the voltage is dynamic - it’s load-dependent. :P The chip generally won’t draw more than it needs at any given moment. We’re talking strictly about peak usage here, but it’s good to narrow things down for everyone.
You can't "draw the voltage you need at any given moment", you get the voltage you are given. You can draw the amps you need from it & you can ask a VRM to provide you more but that isn't what he says.
 
Last edited by smf,

ShiroTheFrog

Member
Newcomer
Joined
Dec 26, 2021
Messages
5
Trophies
0
Age
26
XP
107
Country
Poland
Testing again witcher 3 + OC on my mariko.
switch connected to official charger
commandline with fps & dynares shown
goal - 720p 30fps "locked" - as few DynaRes dips as possible.
area : Ard Skellig - Kaer Trolde near the docks, ~10 npc's, 2 fighting.

Ram 1795.2 at 600mv as typical, i don't bother changing it not to mess anything. Should be fine long term, deliberately left headroom from 1866. Mariko's are much better built than erista.
Cpu - tried to go as low as possible, to save on power limit for gpu. At 1122mhz it's hanging around 84-90% usage, Occasionally hitting 93-94% for split second, possible to go even lower, as 1785 still hovered around 90% use anyways. Weird.
Gpu - well, 720p is absolutely impossible to keep below 1200mhz with dynares on. 1228mhz is the compromise while 1267mhz is ideal. Tiny headroom before hitting 1305mhz hard limit, (i did not test anything above 1267)

921mhz gpu limit can only give you 600-648p range. usually 608-620p. May dip to 26-27fps in big combat casting igni at 10 wolves,
it's still a difference from portable's typical 468p.

oh, i'd forget about temps, although rather short time (<10min) of testing has shown - 44/43/46C for SOC/PCB/SKIN respectively.

i don't have tools to measure the power draw unfortunately. Maybe someone has tested such clocks on mariko & could answer for me. Huge thanks & good luck for Indoo to complete the power draw tests.
 

Foxi4

Endless Trash
Global Moderator
Joined
Sep 13, 2009
Messages
30,825
Trophies
3
Location
Gaming Grotto
XP
29,841
Country
Poland
Trolling as per usual.

I just don't have time to pull apart all your incoherent babble & nobody else can too.

My point is that overclocking causes the gates to switch quicker which requires more amps. The volts are set by what is producing the power. You have a hate boner for me, so you always try to contradict me. But if you have a psu providing 5v you can't suck 6v through because of an increased load.

You literally cannot disagree with this as it's basic physics.

Your wall of text is just pointless for everyone.
It needs more power. I won’t even address your PSU comment, input voltage is orders of magnitude higher than operating voltage, which you’d know if you either read my post or knew anything about the subject. Waste of time.
 

smf

Well-Known Member
Member
Joined
Feb 23, 2009
Messages
6,643
Trophies
2
XP
5,861
Country
United Kingdom
It needs more power. I won’t even address your PSU comment, input voltage is orders of magnitude higher than operating voltage, which you’d know if you either read my post or knew anything about the subject. Waste of time.
Again a non point because you're butt hurt.

Go cry somewhere else.

Switch battery voltage is 3.7v, input voltage is 5V. operating voltage is orders of magnitute lower? Really? That is one low voltage system.
 
Last edited by smf,

3DSBricker

Not So Daily Active Member
Member
Joined
Apr 8, 2020
Messages
281
Trophies
0
Location
A place far away from you
XP
747
Country
Netherlands
I'm not sure I want to go through all his comments... but this is one place.


You can't draw the voltage you need, you get the voltage you are given. You can draw the amps you need from it & you can have something generate a different voltage to give to something else. But that is not what he says.
You're partly right, but Foxi4 is too. It is indeed load dependent. Now we're back at the VID. That's why, it asks for Voltage. But not like it draws it, no, it asks the VRM. The Switch however, is sort of an exception as the Voltage is pre-configured and when the load increases, the Voltage does too. So it can provide enough power. That's why it's dynamic. You're also partly right as "draw power" is a bit confusing if you're like a perfectionist. Still Foxi4 could be talking about current being drawn and not Voltage (which is a vague reference to current though as Foxi4 was talking about Voltage)
 
  • Like
Reactions: Foxi4

Foxi4

Endless Trash
Global Moderator
Joined
Sep 13, 2009
Messages
30,825
Trophies
3
Location
Gaming Grotto
XP
29,841
Country
Poland
This all doesn't belong in this thread anymore. Where did Foxi4 say the Switch draws Voltage? Voltage is supplied by the VRM's and it draws the current. It can, however, ask the VRM to supply a certain amount of Voltage. This is usually called the VID. That's not the point here, though, and neither is it mentioned anywhere.
Thank you, and yes. Voltage on the rail is negotiated between the processor and the controller based on demand and achieved by switching the MOSFET. I believed some clarification was required in regards to how a VRM actually works since we’re talking about various hardware parameters like ratings which the average user may not understand. I think that purpose was achieved already, albeit in a lengthy fashion. Hey, more detail is more better. Tl;dr, voltage is dynamic and set by logic, current just flows, because that’s what it does. Nothing in a VRM circuit negotiates a specific level of current, current exists when there is a load. The chip “consumes” power, not voltage or amps, voltage and amperage are characteristics of electricity within a circuit.
 

Foxi4

Endless Trash
Global Moderator
Joined
Sep 13, 2009
Messages
30,825
Trophies
3
Location
Gaming Grotto
XP
29,841
Country
Poland
Unless there are some real reasons or things we're wrong about and should know, I suggest we stop this (pointless) discussion here. This is getting way off-topic.
Agreed, now that the terms are relatively clear. If Switches are “popping” due to excessive current, it is generally advisable to know why it happens and what causes it. :)
Again a non point because you're butt hurt.

Go cry somewhere else.

Switch battery voltage is 3.7v, input voltage is 5V. operating voltage is orders of magnitute lower? Really? That is one low voltage system.
I would be shocked beyond belief of average operating voltage if the SOC exceeded 1.5V, but I would have to check the X1 datasheet. Someone with more experience probing the rail could probably give us a more specific answer.
 

smf

Well-Known Member
Member
Joined
Feb 23, 2009
Messages
6,643
Trophies
2
XP
5,861
Country
United Kingdom
You're partly right, but Foxi4 is too.
Thing is, what he's right about is stuff that is irrelevant to my original point.

He does it on purpose to derail any conversation I'm involved with and then acts all superior and insults me.

If he wasn't a jerk about it when he tries to correct a point he misunderstood then it would be a lot more pleasant around here.
 

smf

Well-Known Member
Member
Joined
Feb 23, 2009
Messages
6,643
Trophies
2
XP
5,861
Country
United Kingdom
I would be shocked beyond belief of average operating voltage if the SOC exceeded 1.5V, but I would have to check the X1 datasheet. Someone with more experience probing the rail could probably give us a more specific answer.
orders of magnitude higher than 1.5v is at least 150v (you don't specify whether it's 2 or more). Nothing inside the switch could cope with that.
 

Foxi4

Endless Trash
Global Moderator
Joined
Sep 13, 2009
Messages
30,825
Trophies
3
Location
Gaming Grotto
XP
29,841
Country
Poland
Thing is, what he's right about is stuff that is irrelevant to my original point.

He does it on purpose to derail any conversation I'm involved with and then acts all superior and insults me.

If he wasn't a jerk about it when he tries to correct a point he misunderstood then it would be a lot more pleasant around here.
I corrected you because you were wrong. It has nothing to do with our, admittedly unpleasant, past encounters. I’m not petty.
orders of magnitude higher than 1.5v is at least 150v (you don't specify whether it's 2 or more). Nothing inside the switch could cope with that.
If the supply voltage is 3.7V nominal, a peak of 1.5V on the rail is less than half that. On a GPU this is even more obvious since most modern cards operate on 12V input only, and buck down to the required 1-ish volts at the VRM.

Again, no hard feelings. You just seem to be confusing power for current, or vice versa, and a clarification was required.
 

3DSBricker

Not So Daily Active Member
Member
Joined
Apr 8, 2020
Messages
281
Trophies
0
Location
A place far away from you
XP
747
Country
Netherlands
I have no idea what your past encounters are and I don't think it's going to make me neutral in who's right about what. GPU's (at least the desktop and laptop things) have lots of limits so they can't get past (mostly) 1.3V and software only allows up to about 1.2V as constant Voltage and just below 1.3V as peak/boost Voltage. The Switch's regulators only can supply up to about 1.525V. Where "only" is actually quite high. Erista goes up to like 1.2V-1.3V with overvolt (and if clocked very high 1.4V) iirc and Mariko 1.22V with overvolt and about 900-1000mV without. Not sure about non-overvolt Voltage for Erista.
 
  • Like
Reactions: Foxi4

Foxi4

Endless Trash
Global Moderator
Joined
Sep 13, 2009
Messages
30,825
Trophies
3
Location
Gaming Grotto
XP
29,841
Country
Poland
I have no idea what your past encounters are and I don't think it's going to make me neutral in who's right about what. GPU's (at least the desktop and laptop things) have lots of limits so they can't get past (mostly) 1.3V and software only allows up to about 1.2V as constant Voltage and just below 1.3V as peak/boost Voltage. The Switch's regulators only can supply up to about 1.525V. Where "only" is actually quite high. Erista goes up to like 1.2V-1.3V with overvolt (and if clocked very high 1.4V) iirc and Mariko 1.22V with overvolt and about 900-1000mV without. Not sure about non-overvolt Voltage for Erista.
Informative, I wasn’t exactly sure what the operating voltage of the SOC was in either iteration, so this helps with some basic napkin math. I’m not that interested in things that happened in the past myself, and I don’t hold grudges. I wouldn’t write several paragraphs of an explanation on how a VRM works, what components it consists of, what those components do in circuit and how that circuit is regulated if I did, my main objective is to keep the thread relatively informative and dispel various myths. One of the points of contention in the thread was that the average newbie won’t know that changing these settings will have consequences in regards to power delivery, and may cause it to exceed parameters it was designed and rated for. That’s true - it’s not just a number that you pick willy-nilly. As such, a more informative post was prudent, especially if I see claims that I’d consider at best incomplete.
 

Chrisssj2

Well-Known Member
Member
Joined
Feb 12, 2008
Messages
2,704
Trophies
1
XP
4,384
Country
Netherlands
Did anyone test this with an oled switch yet? Is oled switch same as mariko? Get same result? Same powerdraw etc?
 
Status
Not open for further replies.

Site & Scene News

Popular threads in this forum

General chit-chat
Help Users
    K3Nv2 @ K3Nv2: Crowbar?