Download "Влияние таймингов памяти в играх."
%3Aformat(webp)%2Fi.ytimg.com%252Fvi_webp%252FlJqN7-d8O2w%252Fmaxresdefault.webp&w=1139&q=75)
Please wait. We're preparing links for easy ad-free video watching and downloading.
Similar videos from our catalog
%3Aformat(webp)%2Fi.ytimg.com%252Fvi%252FUYvb30VH5K4%252Fmaxresdefault.jpg&w=1139&q=75)
%3Aformat(webp)%2Fi.ytimg.com%252Fvi%252FUxVg7aAJbNY%252Fsd2.jpg%253Fsqp%253D-oaymwEoCIAFEOAD8quKqQMcGADwAQH4AbYIgAKAD4oCDAgAEAEYHyByKBEwDw%253D%253D%2526rs%253DAOn4CLD5BrVibCfnXHYwU-Hnmlu1HDkhmQ&w=1139&q=75)
%3Aformat(webp)%2Fi.ytimg.com%252Fvi_webp%252Fbea7NAyieWc%252Fmaxresdefault.webp&w=1139&q=75)
%3Aformat(webp)%2Fi.ytimg.com%252Fvi_webp%252Fh6Z9CkUbD2A%252Fmaxresdefault.webp&w=1139&q=75)
%3Aformat(webp)%2Fi.ytimg.com%252Fvi%252FxApxT7TmGW0%252Fsddefault.jpg&w=1139&q=75)
Video tags
Subtitles
Russian
00:00:01
we have already talked about the influence of RAM frequencies in games00:00:03
and I highly recommend watching00:00:05
that video before if you haven’t watched it00:00:08
since I won’t tell you the whole theory of memory hierarchy for the second time,00:00:11
and in this00:00:13
video I will assume that you are already00:00:15
completely familiar with what was said earlier00:00:17
in short, the point is that as the00:00:19
memory frequency decreases,00:00:22
processor downtime associated with hunger increases.00:00:24
Understand that the processor00:00:26
faces a task, but it cannot solve it while it00:00:29
expects data from the RAM,00:00:31
and in different circumstances with00:00:33
different settings, the influence is either visible00:00:35
only by the level CPU load00:00:37
or performance in games and00:00:39
in order to understand what’s what, you’ll have to watch the first half of the00:00:41
video about memory frequencies,00:00:43
but let’s return to the main topic00:00:46
of timing, first you need to generally00:00:48
understand what timings are and how they are00:00:50
measured and what the00:00:53
theory suggests Let's see what will happen in games00:00:55
Timings are numbers by which you can00:00:57
judge the speed of access to memory cells,00:00:59
they are all measured in numbers, so in00:01:02
memory, that is, these numbers are like in the last00:01:04
test 1517 1735 is the number of clock cycles00:01:08
necessary to respond to requests from00:01:10
the processor to activate the necessary areas00:01:12
understand the required row of the required00:01:14
memory columns, but the duration of the clock cycle depends on the00:01:17
frequency of the memory, since the higher the frequency,00:01:19
the shorter the clock time itself, but there is00:01:22
another subtle thing that the memory itself operates00:01:24
at half the frequency of what we00:01:26
call clock memory, that is, the00:01:28
frequency that is written in the form of 1600 megahertz00:01:30
for memory at a frequency of 3200 megahertz, this is00:01:34
all for a reason; this frequency is from which exactly the clock cycles00:01:36
for timings need to be counted.00:01:38
Anyone who remembers a school physics course00:01:40
knows that when oscillating, this is the00:01:42
same as a period, and a period is the00:01:44
inverse value of purity,00:01:45
that is 1 divided by frequency is the period00:01:48
for 1600 megahertz00:01:50
the period is 0. six hundred twenty-five00:01:52
thousandths multiplied by ten to the minus00:01:53
eighth power of a second, respectively,00:01:56
we multiply all this by the numbers in the timings00:01:57
and get the time in seconds, more precisely in00:02:00
fractions of a second, of course now we’ll figure out00:02:02
what these delays mean when they00:02:04
occur, consider both values in00:02:06
this case 1500:02:08
ago 35 to understand What do these00:02:11
numbers mean? You need to imagine memory as a set of00:02:13
large tables. In order to access the00:02:15
desired value, you need to know which00:02:17
row of the table to look for and in which column of00:02:19
the table to look for. The first digit indicates the00:02:22
number of clock cycles required to access the00:02:24
value in the desired column.00:02:27
Some physical processes occur in memory, the00:02:29
duration of which it00:02:31
is the delays to find and processes that differ from zero and are00:02:34
set using timings faster00:02:36
than is physically possible, the memory00:02:38
cannot work and, accordingly, if00:02:40
the control is told to wait less than00:02:42
possible, then it will work00:02:44
unstable periodically without producing the00:02:46
necessary data, which in reality00:02:48
is reflected in the form of a blue screen in Windows00:02:51
or simply00:02:52
the computer not working, that is, it is impossible to reduce the timings to 0;00:02:55
this applies to everyone; the number is not only00:02:57
1; the second digit indicates the number of clock cycles00:03:00
required to gain access to the00:03:02
desired line; the third digit indicates the00:03:04
number of clock cycles required to move to00:03:06
another line, that is, to complete work00:03:08
with the previous one readiness to switch to a00:03:10
new one and the fourth digit approximately00:03:12
temporarily reflects the entire reading process and00:03:14
the completion of reading the sequence00:03:16
of timings since it is written00:03:18
does not correspond a little to how they00:03:20
proceed during reading, so the first value00:03:23
is the very last thing that happens and00:03:25
before selecting the desired column00:03:27
you must first select the desired line, that00:03:30
is, the first plus the second value, or00:03:31
go from another line to the desired one, that is, the00:03:34
third value plus then the second plus the00:03:36
first, that is, first there are 2 or 300:03:38
timings, then 1, then to00:03:41
continue working within one line, there are00:03:43
already five only 1 timing within00:03:45
reading on one line, in principle,00:03:47
only 1 timing is important, which is why it is00:03:50
considered the most important and significant.00:03:52
There is also a misconception that the smaller the timings00:03:54
in the form of numbers, the less delay this is not00:03:57
entirely true, the numbers themselves are not an00:03:59
absolute thing; they are expressed not in seconds but00:04:01
in clock cycles,00:04:02
so frequency also affects delays00:04:04
than the purity of the highest with equal numbers of00:04:06
delay timings which delays are less I will00:04:09
conventionally depict two frequencies 1 is 100000:04:12
megahertz and 2 is 2000 megahertz for 100000:04:15
megahertz 10 clock cycles is here for 200000:04:19
megahertz here and00:04:20
clock cycles are half as many, which is why00:04:22
with increasing frequency there is an00:04:24
increase in the value of timings since00:04:26
for different frequencies the physical processes00:04:29
occurring in memory do not change and00:04:31
require the same time and in fact,00:04:33
over the last 30 years, those00:04:35
physical processes in memory have accelerated by not even an00:04:37
order of magnitude, that is, less than 10 times, which is why00:04:39
even purity has increased00:04:41
many times the timings are expressed in seconds;00:04:44
they decreased in clock cycles by only a00:04:46
few times; however, considering that00:04:48
not all types of reading00:04:50
involve timings in the process, in practice the00:04:52
increase in memory speed has increased00:04:55
significantly. I suggest you look at00:04:57
Wikipedia here for a table for00:04:59
memory latency during sequential reading for the00:05:01
first reading for a memory of 100 megahertz with00:05:04
timings of 2 requires 20 nanoseconds for00:05:07
memory of 2000 megahertz with a timing of 10 1000:05:10
nanoseconds,00:05:11
that is, with a 20-fold increase in00:05:12
frequency, the speedup was only00:05:16
twofold due to large timings, but00:05:18
if you look further when reading in00:05:20
order on the eighth request at 200000:05:22
megahertz 13.5 nanoseconds will be spent,00:05:26
while at 100 megahertz00:05:28
90 for seconds, that is, the difference is00:05:31
almost seven times and you can also see how00:05:33
important timings are at low frequencies,00:05:35
let’s say for 200 megahertz of memory the difference00:05:39
between timings 2 and 300:05:40
to the eighth word was five00:05:43
nanoseconds and the difference between two and three is00:05:46
50 percent if we count the00:05:48
smaller value as one hundred percent, and for 200000:05:50
megahertz the difference between timings 9 and 1000:05:53
to the eighth word was one and a half00:05:55
nanoseconds the difference in timing between00:05:58
nine and ten is only 11 percent and00:06:01
in my opinion it is absolutely obvious that the00:06:03
real significance of timings after the00:06:05
concept of theory is completely unobvious without00:06:08
real tests in games,00:06:09
that is, if I conducted a video about00:06:12
frequencies there were three series of tests, the first00:06:14
was where the processor was not fully loaded,00:06:16
2 when the processor was fully loaded and 300:06:19
when caching of video00:06:20
data occurred RAM and the processor00:06:22
was not fully loaded in the first00:06:24
case, we saw that at00:06:26
fps the percentage of processor load the leader00:06:28
was displayed higher, in the theoretical part of00:06:30
that video before that I explained why this00:06:33
happened with the timings, the same thing00:06:35
would have happened in the test with the processor00:06:37
not fully loaded I in this video I didn’t do it, it’s00:06:39
obvious that the essence of the results would be00:06:42
approximately the same, that is, there are00:06:44
only two remaining cases, and I also00:06:46
didn’t film the tests themselves, this is00:06:48
all quite inconvenient for me since the00:06:50
required accuracy does not allow00:06:52
recording with software capture and00:06:54
I have to remove the monitor camera and it is00:06:56
necessary to set up the camera to monitor the00:06:58
light, I would not have found free time for this00:07:00
and the video would not00:07:02
have come out at all. Settings in games also do not00:07:05
matter at all, the main thing is that with different00:07:07
timings they were all the same; all00:07:09
conditions in the tests were identical except for00:07:11
themselves timings, the test bench00:07:13
consists of a 777 20 ok00:07:15
which worked at a frequency of 3 and 500:07:17
gigahertz with hyper trading disabled00:07:19
to make it easier to load the processor 1600:07:21
gigabytes of RAM from a year ago00:07:23
two strips of 8 gigabytes each for tests,00:07:26
two frequencies were set, the00:07:27
first is the base frequency for ddr4 213300:07:31
megahertz with timings 12 12 12 28 as the00:07:36
smallest value that00:07:37
word memory allows me in Russian retail00:07:39
and I have not seen similar plans, and where00:07:42
I managed to find similar memory with00:07:44
such timings it turned out that00:07:46
it costs even more than good frequency bars,00:07:49
that is, 12-12 1228 is a pretty00:07:52
ideal option for ddr4,00:07:54
then at the same frequency00:07:56
15-15 1535 was set as the most00:08:00
common strips on sale and 171700:08:03
1742 as the worst that I could00:08:06
find on sale00:08:07
and also for comparison the00:08:08
frequency of another 3200 megahertz and timings00:08:12
1517 1735 were set this is the most productive00:08:15
mode of operation of my memory that I managed to00:08:17
achieve; comparisons will be00:08:19
carried out at a higher frequency separately, that is,00:08:21
after the main tests of video cards, an00:08:23
rx 470 was installed at 4 gigabytes for00:08:26
tests in the second case, when00:08:28
it was necessary to load all the video memory,00:08:30
the processor worked at 4 4 8 gigahertz in00:08:33
full that is, four cores and an00:08:35
enabled hypertrading00:08:36
video card,00:08:38
it was up to a frequency of 1400 megahertz on the00:08:40
chip, also for tests using00:08:42
RAM to polarize00:08:44
video data in a test where I tested the00:08:47
strips at different frequencies, you asked not to00:08:49
use gta5 since it leaves00:08:51
a lot of video memory space for emergency00:08:54
loading anything, although there is a video about the00:08:57
restriction of the piss express where you can see that it still00:08:59
runs, it doesn’t help with00:09:01
caching into the RAM, and you also00:09:03
asked for tests with Ovz and there is a raider test in00:09:06
Lara Croft for a video about frequencies,00:09:08
I also actually have that the video00:09:10
that I left with the tests is at the link in the00:09:12
description of the video about memory frequencies, part of the00:09:14
game and then I didn’t take the final video00:09:16
to tame the video to at least 30 minutes00:09:19
for all tests I saved their plays00:09:21
frame time and so in the first case00:09:23
the processor was fully loaded the first game00:09:25
in watch dogs 2 test location00:09:27
has to be done manually, also on the roads there is00:09:29
always a different amount of traffic00:09:32
so the data may be a little inaccurate,00:09:34
I just pass on motorcycles sometimes I00:09:36
come across faster sometimes00:09:38
slower, which introduces certain00:09:40
problems with comparison visually according to the00:09:42
graphs, passing in the second half00:09:44
of the tests do not coincide at all in the first00:09:46
half it is clearly visible that low timings00:09:49
have a positive effect on00:09:50
performance, you are asking me to do, in addition to00:09:52
graphs, just numbers00:09:54
separate saw ps here are the numbers average fps00:09:57
fluctuates within the margin of error00:09:59
the minimum is probably also not without00:10:01
some errors but the difference between twenty00:10:04
seven or thirty three fps per00:10:06
error It is clearly not possible to write off the00:10:08
frame time graph for the first half, it is clear00:10:10
that with lower timings the00:10:12
frame time is clearly lower, the stability of the frame time with00:10:14
timings 17 is clearly worse than Crysis 3,00:10:17
the testing was carried out at the end not the00:10:20
same as in the memory frequency test, the difference00:10:22
from the timings in the game is also not high and All00:10:25
data generally fits within the limits of00:10:27
testing errors according to00:10:29
frame time graphs, also strong;00:10:31
no significant difference is visible; gta 5 presents00:10:34
data on the last scene of the benchmark; the one00:10:36
that starts with an airplane and then the00:10:38
next car in the city; benchmark00:10:41
gta5 has some variations in the number of00:10:43
cars in the city00:10:44
so his data is also not00:10:46
the prices based on dry figures, it’s difficult to say00:10:48
how much the increase in fps is higher than the00:10:50
errors; the minimum 50 and 52 frames,00:10:54
given the specifics of the benchmark, are not00:10:55
indicative; however, the difference between the00:10:58
average fps is clearly higher than the errors, which00:11:00
confirms the frame time graphs;00:11:02
it is clearly visible that there is a difference between00:11:04
12 and 15 and an even more significant difference00:11:08
between 15 and 17 you, while the frieze00:11:11
shows nothing in terms of differences, then00:11:13
hitman00:11:14
testing was carried out in the00:11:16
game benchmark, I liked this benchmark, this one00:11:18
produces quite stable numbers, so the00:11:20
difference is even between 50 and three and 5000:11:23
five fps errors cannot be written off00:11:26
within the framework of tests in one00:11:28
system, the increase from reducing timings is00:11:30
definitely there, unfortunately, due to the00:11:32
peculiarity of recording the time frame not from00:11:34
the screen but at the input to the video card, graphs00:11:36
with frame time cannot be analyzed in any way00:11:38
for clarity, I present the00:11:41
fps graphs clearly visible that green is below00:11:43
and blue in the middle and red on top the00:11:46
next game NFS 15 there is no benchmark in the game the fact00:11:49
that I am going through the same00:11:51
race in the game trying to repeat the00:11:52
passage this time unfortunately on00:11:54
one of the three passages I was somewhat lost00:11:56
in time so on the graphs I can00:11:59
only show timings 12 and 17, it’s clear that the00:12:02
difference is not radically high, but still00:12:04
predominantly red has less00:12:06
frame time. As for the obtained00:12:09
PFTs figures, the00:12:10
average value is quite within the00:12:12
errors, given the lack of a00:12:14
benchmark,00:12:15
but the minimum fps has clearly increased 1200:12:18
frames per second or twenty two point00:12:20
six percent cannot00:12:22
be written off in any way errors as for the00:12:24
average for 5 games you00:12:26
got the following result when switching from00:12:28
timings 17 to 15 I received an increase in00:12:31
average fps of about 2 percent, at a00:12:34
minimum of about 3.8 percent when00:12:37
switching from 17 to 12 received an increase in00:12:40
average fps of about 2.8 percent and in00:12:43
minimum fps almost 11.800:12:46
percent, the data is not very clear,00:12:49
of course, when there00:12:50
is no more time, these tests00:12:53
will need to be done again in a larger00:12:55
number of games and immediately running all the tests00:12:57
several times so that To exclude00:12:59
errors, the video will already be without theory,00:13:01
that is, more developed practical00:13:04
tests, at the moment00:13:06
I don’t have such capabilities and it’s not possible to see the exact00:13:08
values from these data,00:13:10
however, when the processor is fully loaded, the00:13:13
timings have an effect in games, this is00:13:15
absolutely clear now let’s move on to the00:13:18
second test where our00:13:19
video memory should be completely filled and00:13:22
caching into RAM should occur, there00:13:24
will only be two games, the first is00:13:26
raises from this raider with ultra00:13:28
texture settings and watch dogs 2 with00:13:30
maximum texture quality and 4k00:13:33
resolution let's look at that raider here is00:13:36
data from the latest benchmark scenes, since it is00:13:38
in this scene that the RAM load00:13:40
reaches 10 gigabytes when00:13:42
there are four gigabytes of video memory in00:13:44
the video card, by the numbers the difference is extremely00:13:46
insignificant, it can definitely be attributed to an00:13:49
error and according to the frame time graphs00:13:51
also the absolute parity of watch dogs 2 the00:13:54
situation is absolutely similar according to the00:13:57
fps data the difference frames that are low in time are negligible00:14:00
and are absolutely not visible, in other00:14:02
words, exactly the same conclusion from this00:14:05
experiment as in the video about00:14:07
memory frequencies: RAM has little effect on00:14:09
performance when caching00:14:11
video data since there is a narrow bus express on the way00:14:13
relative to other00:14:16
buses between the RAM is a video chip and00:14:19
there are also several controllers on the way on00:14:21
which the delays apparently00:14:22
increase greatly so that the timings themselves00:14:25
no longer solve anything, and a cycle of two videos00:14:27
would be incomplete if we did not conclude that the00:14:29
frequency or00:14:32
timings on the device decide more in fact, this whole concept is00:14:35
not mutually exclusive but complementary,00:14:38
so a high frequency will not give an increase00:14:40
if reading requires a lot00:14:42
more clock cycles and the timings will not give00:14:45
any profit if they do so, they will be00:14:47
very long, so overclocking00:14:49
RAM is a search for the00:14:51
optimal operating mode when I have a00:14:53
high frequency and at the same time, for00:14:55
normal operation, moderate00:14:57
timings are available for my memory, which was00:15:00
the most productive mode in the test,00:15:02
being00:15:03
at a frequency of 3200 megahertz with secret 151700:15:06
1735, as you have already seen, testing in00:15:09
games is not the most convenient method, I00:15:12
recommend that when overclocking memory, in addition to00:15:14
stability tests,00:15:15
measure the performance of the subsystem00:15:16
memory using benchmarks in00:15:18
archivers, there is a built-in benchmark00:15:21
in both winrar and 7-zip,00:15:22
so each of you already has the00:15:24
necessary tools for00:15:26
memory tests during overclocking, but I would recommend the00:15:28
winrar test, its load is00:15:31
a little more theoretical and depends00:15:33
on memory it shows above00:15:35
actually for all tested00:15:36
memory operating modes the00:15:38
numbers turned out winrar like this and00:15:41
of course for 1 goal and theories about cleanliness00:15:43
and timings I won’t stop because in addition to00:15:46
tests at a frequency of 2133 megahertz I00:15:49
did another set of tests for the first00:15:51
case when loaded the processor is00:15:53
completely at a frequency of 3200 megahertz00:15:56
with timings of 1517 1735 and now on all the00:15:59
graphs and diagrams I will add data from the00:16:01
torn memory in the blood of seo 5 nothing00:16:04
sensible can be seen not by the fps value not00:16:07
by the frame time in gta 5 by the numbers again it’s00:16:10
not clear what’s what It’s clear that with00:16:13
memory overclocking, the average FPS has grown quite a bit; the00:16:16
drawdowns aren’t all that clear, but as00:16:18
I said ted in gta5 is not very good,00:16:21
but looking at the graphs with frame time,00:16:23
it’s obvious that overclocking the frequency00:16:26
greatly increases the performance in the00:16:28
game in Hitman, that’s all true the00:16:31
final data clearly shows the increase from00:16:34
overclocking in comparison with timings 17 at a00:16:36
frequency of 2133 megahertz is extremely00:16:40
significant and exceeds, for example,00:16:42
the transition or even five to 7.00:16:43
As I already said, the frame time graphs00:16:46
in this game do not work out, but the fps graphs are00:16:49
amazing the difference from overclocking it’s just00:16:51
huge in NFS 15 the difference is even greater,00:16:55
unfortunately, since the passage does not00:16:57
converge exactly on the graphs, this difference is00:16:59
not so easy to feel, but if you find the peaks00:17:02
and valleys, which are the same for all00:17:04
passages in the00:17:06
same places but just don’t coincide a little00:17:08
in time, that is, the lens is shifted, it is00:17:10
also very clearly visible watch dogs 2, the00:17:13
increase in frequency is manifested both on00:17:15
average and in the minimum FPS, while in00:17:18
this game the difference is a fundamental00:17:20
drop in FPS became above 30 and the average00:17:24
increased to 50, which is strikingly different in terms of00:17:26
comfort from memory with00:17:28
timings are 17, but we also don’t forget that00:17:31
as I showed her in the video about the memory frequency,00:17:34
if the percentages are fully loaded, then the00:17:36
difference will not be visible at all, so00:17:38
I repeat once again that overclocking the memory is not00:17:40
magic that will make a super powerful00:17:43
computer a super mega powerful computer, but00:17:46
if you let’s say midi is a weak processor,00:17:48
then memory overclocking is a very useful thing00:17:51
if it is available on your motherboards. Also,00:17:53
memory overclocking is very necessary if for full00:17:55
hd00:17:56
you have a video card oriented00:17:58
towards higher resolution, that is, the current00:18:00
top00:18:01
cards of course for 4k full hd give out a00:18:04
potentially very large dog and in00:18:07
order to the processor met00:18:08
the requirements; you need to reduce all its00:18:11
downtime, including memory starvation, and this will00:18:14
also increase performance. Well, as an00:18:16
obvious conclusion, if you can00:18:18
seriously increase the performance of00:18:20
the frequency,00:18:21
then modern realities with ddr4 sacrifices in the00:18:24
form of two or three delay cycles are not so00:18:27
critical, and also pay attention00:18:29
to timing when buying plans since with00:18:32
15 17 they cost almost the same and00:18:34
in terms of performance in some00:18:36
moments they will differ, well, as I00:18:38
promised earlier, we will continue with you and in the00:18:40
future the topic of work, understand, I briefly00:18:43
told a video about cleanliness and about the00:18:44
memory hierarchy and also It’s worth mentioning once again00:18:47
that she has a son who connects the00:18:50
processor cache with the cores and she also00:18:53
causes downtime of the processor when all00:18:55
its tasks require handling the mess and in00:18:58
this video the frequency and you of the bus and of course the00:19:00
processor were unchanged in Ryzen, to00:19:02
this another bus is added through00:19:04
which an array of four cores00:19:06
accesses the cache from the other four cores in an00:19:09
8-core processor and, accordingly,00:19:11
three cores in a 6-core processor, and this00:19:14
also affects delays and00:19:16
limitations on system performance, so the00:19:19
memory issue is still far from00:19:20
complete, but I think with the00:19:23
operating frequency memory and timings, everything is00:19:24
more or00:19:25
less clear if the video was useful then00:19:28
don’t forget about likes and also00:19:30
subscribe to the channel if you have00:19:31
n’t done so before, but for now everyone, bye until00:19:35
new videos00:19:36
[music]00:19:46
[music]Description:
Видео про частоты памяти: https://www.youtube.com/watch?v=xmUQFl25Ry0 https://pc-01.tech/ - сайт канала. Свежие новости о железе, обзоры и тесты комплектующих. В видео мы разбираемся что такое тайминги оперативной памяти, и тестируем в играх систему с разными таймингами. https://vk.com/pc_0_1 - группа "Этот компьютер" - свежие и актуальные новости IT мира
Preparing download options
* — If the video is playing in a new tab, go to it, then right-click on the video and select "Save video as..."
** — Link intended for online playback in specialized players
Questions about downloading video
How can I download "Влияние таймингов памяти в играх." video?
http://unidownloader.com/ website is the best way to download a video or a separate audio track if you want to do without installing programs and extensions.
The UDL Helper extension is a convenient button that is seamlessly integrated into YouTube, Instagram and OK.ru sites for fast content download.
UDL Client program (for Windows) is the most powerful solution that supports more than 900 websites, social networks and video hosting sites, as well as any video quality that is available in the source.
UDL Lite is a really convenient way to access a website from your mobile device. With its help, you can easily download videos directly to your smartphone.
Which format of "Влияние таймингов памяти в играх." video should I choose?
The best quality formats are FullHD (1080p), 2K (1440p), 4K (2160p) and 8K (4320p). The higher the resolution of your screen, the higher the video quality should be. However, there are other factors to consider: download speed, amount of free space, and device performance during playback.
Why does my computer freeze when loading a "Влияние таймингов памяти в играх." video?
The browser/computer should not freeze completely! If this happens, please report it with a link to the video. Sometimes videos cannot be downloaded directly in a suitable format, so we have added the ability to convert the file to the desired format. In some cases, this process may actively use computer resources.
How can I download "Влияние таймингов памяти в играх." video to my phone?
You can download a video to your smartphone using the website or the PWA application UDL Lite. It is also possible to send a download link via QR code using the UDL Helper extension.
How can I download an audio track (music) to MP3 "Влияние таймингов памяти в играх."?
The most convenient way is to use the UDL Client program, which supports converting video to MP3 format. In some cases, MP3 can also be downloaded through the UDL Helper extension.
How can I save a frame from a video "Влияние таймингов памяти в играх."?
This feature is available in the UDL Helper extension. Make sure that "Show the video snapshot button" is checked in the settings. A camera icon should appear in the lower right corner of the player to the left of the "Settings" icon. When you click on it, the current frame from the video will be saved to your computer in JPEG format.
What's the price of all this stuff?
It costs nothing. Our services are absolutely free for all users. There are no PRO subscriptions, no restrictions on the number or maximum length of downloaded videos.