Is NoPixel V dead? No, but its audience fell 82%

Update, 15 October 2026 — three days later it is back, and above launch day. In the same clock window, across eight complete censuses between 04:10:36 and 04:17:19 UTC, NoPixel drew a mean 127,364 concurrent viewers on 144 channels — 49.7% of all live GTA viewing, the highest share we have recorded for any single roleplay server. Day one, with Twitch paying, was 125,031 and 39.2%. The 20,212 below stands as measured, but what it caught was a trough rather than a floor: live GTA at 04:00 UTC on 12 October was 116,908, the lowest of 31 daily readings in our thirty-day window, against 261,656 this morning, eighth of 31. The structural argument further down this page is the part that held — an audience resting on a handful of very large channels can halve when they do something else, and it can come back the same way. xQc, Jerma985 and buddha were all live on V this morning. We have no confirmed cause and make no causal claim.

No. This morning NoPixel drew a mean of 20,212 concurrent viewers across Kick, Twitch and YouTube — more than four times the next-biggest roleplay server, and roughly thirty times what NoPixel drew six days before its Rockstar-published version existed.

It is also down 82.6% from where we measured it three weeks ago.

Both of those are true, and the gap between them is the whole story. The number people remember is the one from the launch fortnight, when Twitch was paying GTA$1.5m in in-game currency to anyone who left a stream open. That campaign ended on 30 September. This is the first measurement of what NoPixel looks like without it.

What we measured this morning

Twelve complete censuses of every live Grand Theft Auto stream on Kick, Twitch and YouTube, taken between 04:07:41 and 04:18:41 UTC on 12 October 2026. Live GTA across that window ran a mean of 124,861 concurrent viewers on 1,758 channels.

	NoPixel, 12 Oct 2026
Mean concurrent viewers	20,212
Range across the 12 censuses	17,674–21,800
Distinct channels	55
Share of all live GTA viewing	16.2%
Share of everything our classifier files as roleplay	40.4%
Platform split	Twitch 75.1%, Kick 24.9%
Largest single channel’s share	20.8%
English-language share	100%

12 censuses, 04:07:41–04:18:41 UTC, 12 October 2026. Streams matched by server name in the title.

That range needs explaining, because a 19% swing inside eleven minutes looks like instrument noise and is not. For the first eight censuses NoPixel sat in a tight band of 20,904 to 21,800. At 04:15:43 three broadcasts ended at once — GEEGA on both Twitch and Kick, and Bazza on Twitch — and the total dropped to 17,872, where it stayed. Three channels closing took 15% of a server’s audience with them in a single minute. Hold that thought.

The shape of it

Share is the right axis here and raw viewers is the wrong one. Live GTA viewing at 04:00 UTC has itself fallen about 17% over the past month on a seven-day mean, and this morning’s reading was the lowest 04:00 point in our entire thirty-day window. Quote the raw fall and you are partly quoting a quiet Sunday. Quote the share and the category’s size cancels out: NoPixel held 36.2% of live GTA viewing on 22 September and holds 16.2% now. It has lost a little over half its grip on the category in three weeks.

Why it fell, and the part that is confirmed

Twitch’s own blog is unambiguous about the dates. GTA RP Week Drops ran 8–30 September and paid GTA$250,000 for one hour watched, another GTA$250,000 at two hours, a further GTA$1m at four, and a Burger Shot Tracksuit at six — up to GTA$1.5m for leaving a tab open. (Confirmed by Twitch.) They were a Twitch mechanic; Kick and YouTube viewers were never eligible.

If a meaningful part of the launch audience was there for the currency, the clearest fingerprint would be Twitch’s share of NoPixel viewing falling once payment stopped. It has, modestly: 86.2% on 9 September, 80% on 22 September, 75.1% today. That is a real eleven-point drift in the direction the theory predicts, and it is much too small to carry the whole 82.6%. Most of what left was Twitch viewers, but most of Twitch’s NoPixel audience did not leave.

The larger mechanism is the one the 04:15 drop illustrates. NoPixel’s launch number was never broad — it was a handful of very large channels. On day one xQc alone held about 46% of it across his simultaneous Twitch and Kick broadcasts. Of the ten biggest NoPixel streamers we measured on 9 September, nine were not streaming GTA at all during this morning’s window, on any server. Only Sykkuno was live, and he was playing freeroam.

Be careful with that sentence, because it is weaker evidence than it looks. These are mostly North American streamers and 04:00 UTC is around midnight on the US east coast on a Sunday. Nine absences in one eleven-minute window says very little about anyone’s week, and nothing at all about anyone’s intentions. What it does establish is structural: an audience assembled on top of ten channels is an audience that can halve because those ten channels did something else, and that is a different fragility from people losing interest in a server.

The concentration has genuinely loosened, though. The largest single channel is 20.8% of NoPixel’s audience today against xQc’s 46% on day one. What survived the Drops is spread more evenly than what arrived with them.

What it is not

It is not a collapse into irrelevance, and the league table is the check on that. Same censuses, same method, every GTA roleplay server we can name from stream titles:

Server	Mean viewers	Channels	Platform split
NoPixel	20,212	55	Twitch 75%, Kick 25%
The Streets RP	4,537	15	Kick 98%, Twitch 2%
ChaseRP	4,099	26	Twitch 100%
Final Chapter	3,712	31	Kick 90%, Twitch 10%
DovuX Life RP	3,451	25	Kick 100%
Headliner RP	2,971	75	Twitch 89%, Kick 11%
FamilyRP	2,290	10	Kick 100%
The Palms	1,442	15	Kick 95%, Twitch 5%
Complexo (CPX)	1,410	29	Kick 92%, Twitch 8%
Prodigy RP	837	40	Kick 72%, Twitch 28%
District 10 (D10)	778	17	Kick 64%, Twitch 36%
South Beach	709	2	Kick 99%, Twitch 1%
Prestige RP	581	22	Twitch 59%, Kick 41%
The Academy RP	391	8	Twitch 100%
Cidade dos Anjos	103	14	Twitch 79%, Kick 21%
New Day RP	23	2	Twitch 100%

12 censuses, 04:07:41–04:18:41 UTC, 12 October 2026. Titles naming two servers are counted in both rows.

NoPixel is 4.5x the server beneath it and holds more viewers than the next four combined. Before Rockstar’s version launched it was nowhere near that: on 2 September, six days out, a hand-audited count put NoPixel at 639 concurrent viewers, about a third of one per cent of live GTA, far behind a Kick server called Windy City on roughly 12,000. Post-Drops NoPixel is a smaller thing than the launch headlines, and a much larger thing than what preceded them.

One row in that table deserves a flag rather than a reading. Windy City is absent, and we are not publishing a number for it, because the two titles that matched our alias for it this morning are a character backstory and a channel announcing “Windy Soon…”. The server appears to be between seasons, and several creators who named it in September now name Final Chapter or The Streets RP instead. A league table taken on one morning is a snapshot of a carousel: the English-language Kick roleplay scene rotates servers on a season cycle, which is a real limitation on every comparison in this piece that is not about NoPixel.

Grading our own prediction

On 2 September, before any of this, we wrote down what the test would be. It is worth quoting because it was specific enough to be wrong:

The honest test is not the launch-day peak. It is what the number looks like after 30 September, when the Drops stop.

We said, 2 September	Verdict
“NoPixel V will post a Twitch number far above 639”	Right, and still right — 15,179 on Twitch alone today, against 639 across all three platforms then
“A large share of it will be Drops watch time rather than interest in the server, and from the outside those two are indistinguishable”	Partly settled. 82.6% of the peak left when the money did. We still cannot separate a bored tab from a Drops tab, but we can now price the combination
“The honest test is… what the number looks like after 30 September”	Answered here. 20,212, or 16.2% of live GTA

We also promised, in that post, to “re-run exactly the measurement… again in October, and publish what changed… including if the answer is ‘very little’.” The answer was not very little, and this is the re-run.

Re-graded on 15 October. The third row has an answer that moves: three days after this post, the same measurement in the same window read 127,364 and 49.7%. “What the number looks like after 30 September” turns out not to be a number but a range, 20,212 to 127,364 inside four days, and a single morning is a weaker test than we treated it as. The honest verdict on that row is now answered, then unanswered — the Drops ending did not settle where this server sits, because the thing that moves the figure most is which large channels are awake.

The second row is the one we got least right, in a way worth owning. We framed Drops watch time and genuine interest as permanently indistinguishable. Turning the campaign off was always going to be a natural experiment, and we did not say so at the time.

What this does not show

We match server names in stream titles, so every figure here is a floor. A NoPixel roleplayer who writes only their character name is invisible to us, and 17,415 of today’s 20,212 explicitly name V or 5 — the remainder say only “NoPixel” and could be on 4.0 or Public. Read the headline as the NoPixel family.

This is one clock hour. GTA viewership swings several-fold across a day and 04:0x UTC is the small hours in North America, where most NoPixel roleplayers live, so this window undercounts NoPixel rather than flattering it. The 9 and 22 September figures were taken in the same window, which is the only reason the comparison is worth anything.

We count concurrent viewers, not people, and not players. NoPixel V remains invite-only as of today, with roughly 450 slots reported and no public date for wider access (reported, not confirmed by Rockstar). A server that cannot admit more players cannot grow its audience the way an open one can, and a viewership figure says nothing about how full it is.

The 2 September 639 was hand-audited; every other figure here is a raw title match. The 09-02 count discarded roughly half the streams that said “NoPixel” and today’s does not, so the 639 is the most conservative number on this page and the comparison against it flatters the later dates. We would rather publish the asymmetry than quietly audit only the figures that suit the story.

And we have no causal claim. The Drops ending, the novelty wearing off, and ten large channels moving on all point the same way on the same timeline, and one measurement three weeks apart cannot separate them.

Check us

Everything above comes from twelve complete censuses — the live board plus the tail, publish timestamps matched across both, so channel counts and viewer sums reconcile exactly against the platform summary in all twelve, with no channel counted twice — taken 04:07:41 to 04:18:41 UTC on 12 October 2026.

If you want to quote any of it, the permalink is https://luciastream.com/blog/is-nopixel-v-dead/, the sampling window is the one above, and the collection method is documented at luciastream.com/methodology. The underlying numbers are the same ones behind our live board, sampled every 60 seconds, and the leaderboard carries the thirty-day version — both are where you can check whether any of this still holds by the time you read it.

We will run it again. The next honest test is whether 16.2% is a floor or a landing, and that needs a December reading, not an October one.