I ran GTA VI through an RTX 5080 on my PC. Look at this picture. It is f*cking incredible.
On every PS5, the Pro included, the game is locked to 30 fps. On my PC the same frames run at 60 with DLSS 5 lighting. That is the bottom half of the video.
Here is what I measured and how to set it up on yours in 5 minutes.
What the test showed:
> In 80 seconds of gameplay the console put out 2,351 real frames. My PC put out 4,700.
> DLSS 5 takes 4.1 ms per frame and frame generation takes 1.5 ms. The GPU sits at 39%.
> 60 fps holds in real time, with 56 in the worst 1%.
> Half the frames in a PS5 stream are copies. The game renders 30 and the console sends 60.
And it is not just me:
-> Rockstar locks GTA VI to 30 fps on every console, PS5 Pro included (Digital Foundry).
-> Veyra, the app all of this runs on, has 112,000+ downloads (GitHub).
Can you put DLSS 5 on the PS5 itself? No.
The PS5 has an AMD GPU with no NVIDIA cores, and a jailbroken console will not run GTA VI because it needs the latest firmware.
But the PS5 can stream to a PC, so you do not need a $200 capture card.
How to set it up on yours:
Step 1 -> Paste the prompt from the first comment into Claude Code or Codex. The agent checks your GPU, installs Veyra from my repo and sets the GTA VI preset.
Step 2 -> On the PS5 turn on Remote Play: Settings > System > Remote Play.
Step 3 -> In the same menu hit Link Device and type the 8 digits into Veyra.
Step 4 -> In Veyra set PS5 streaming to 1080p, 60 fps, H.265.
Step 5 -> In Pro mode turn on NR "RTX 50 · NVIDIA original" with 2 layers and Frame Gen DLSS 2X, or 4X on a 120 Hz screen.
Step 6 -> Go fullscreen and hold V to see what the console alone looks like.
It works on RTX 40 and 50. DLSS 5 needs an RTX 50.
Controls still react at 30 fps. The picture is what gets smoother.
Do it before your unlock tonight. Prompt and repo are in the first comment.
How people are cashing in on this launch is below ↓
I pointed a Composal Spark at one of my repos and asked for a review.
What came back:
Cloned the repo on its own cloud computer -> Found 3 real bugs -> Wrote a fix for each -> Ran the tests
All 25 passed, and nothing got pushed without me.
That was one agent on one task. The product is a set of agents the whole team shares:
> One picks up a Sentry error and opens a PR with a fix.
> Another reviews it and merges once approved.
If your team already codes with AI and still chases reviews and merge conflicts by hand, the waitlist is at composal.com/sparks.
What is the one bit of engineering work you wish was just automated?
Viktor retweeted
Replying to @xQc_dailyy
Looks to be this software. Seems interesting if it actually works well and doesn't introduce any bugs.
github.com/Likely7/Veyra-NRV…
Viktor retweeted
BP blues but can't stop buidling
updated to both pipeline and the universal bonding curve both pushed
github.com/MetalLegBob/pipel…
github.com/MetalLegBob/pipel…
Viktor retweeted
deployed a bonding curve for Pipeline that should be flexible enough to graduate into most different amms to help people building etc
lmk if you run into issues or need new functions will get sorted
pls don't deploy it to mainnet it's likely got bugs and haven't run any audit passes on it yet either
moga
github.com/MetalLegBob/pipel…
Who wants GTA6 and a PS5?💰
To enter:
• RT
• Follow @BagsApp
• Drop your Bags link below 👇
Winner picked randomly on Friday.
*must be holding at least one GTA related coin in your wallet on @BagsApp to win
bags.fm/@finnbags
Viktor retweeted
We've made some changes to the internal arbitrage bot to keep ZIG and ZAG prices more uniform across their respective sets of LPs.
Although this slightly impacts our growth rate from external volatility, it captures far more spread from internal volume which would otherwise leak to arb bots.
This should provide a better trading experience/less slippage for both buyers & sellers.
Plenty of unexpected issues when you're innovating, we'll endeavour to fix them all 🧪🫡
Viktor retweeted
A rare opportunity to get into a token that’s going to usher in a new era on Solana
@fffactoryhq
One of the biggest blockers to experimentation and innovation on Sol is distribution (or lack of).
Put simply, if your idea doesn't fit inside the box given to you by the big infrastructure (routing, indexing, etc), then no one will find your token/idea/launchpad etc.
Some have tried to counter this by launching a visible token (i.e via pump or wherever) and then directing that attention to their own site but that's still too high friction.
Both the expense of building, and the skill required to do so have been reduced 90%, and we should be using this to innovate and experiment. Not just rely on a few big corps to decide what features are 'worth it'.
So i'm building Pipeline (it will never have a token, and any potential revenue if it succeeds will be part of the @fffactoryhq ecosystem).
Pipeline is a permissionless onchain router for new bonding curves, AMMs, launchpads etc. Creators can register their contracts with Pipeline, and infra (trading bots, terminals, charting) that has Pipeline integrated will pick them up automatically.
Anyone that's interested in experimenting with the platform, looking to build something themselves, whether that's things to register with Pipeline, or infra built on top of it, please drop me a dm or comment here.
Currently only live on devnet as still running tests etc, will likely build a few fun AMMs/T22 hooks that can be experimented with myself to go with the mainnet launch, but anything extra built by others would be awesome (if you build your own amms, bonding curves, launchpads etc you can charge your own fees on them).
github.com/MetalLegBob/pipel…
Viktor retweeted
Started a small tg group for anyone that wants to experiment/build things on pipeline please lmk if you wanna join
make onchain great again
One of the biggest blockers to experimentation and innovation on Sol is distribution (or lack of).
Put simply, if your idea doesn't fit inside the box given to you by the big infrastructure (routing, indexing, etc), then no one will find your token/idea/launchpad etc.
Some have tried to counter this by launching a visible token (i.e via pump or wherever) and then directing that attention to their own site but that's still too high friction.
Both the expense of building, and the skill required to do so have been reduced 90%, and we should be using this to innovate and experiment. Not just rely on a few big corps to decide what features are 'worth it'.
So i'm building Pipeline (it will never have a token, and any potential revenue if it succeeds will be part of the @fffactoryhq ecosystem).
Pipeline is a permissionless onchain router for new bonding curves, AMMs, launchpads etc. Creators can register their contracts with Pipeline, and infra (trading bots, terminals, charting) that has Pipeline integrated will pick them up automatically.
Anyone that's interested in experimenting with the platform, looking to build something themselves, whether that's things to register with Pipeline, or infra built on top of it, please drop me a dm or comment here.
Currently only live on devnet as still running tests etc, will likely build a few fun AMMs/T22 hooks that can be experimented with myself to go with the mainnet launch, but anything extra built by others would be awesome (if you build your own amms, bonding curves, launchpads etc you can charge your own fees on them).
github.com/MetalLegBob/pipel…
Viktor retweeted
I love Bob.
I love what he’s building.
Well worth a read.
Be more like Bob.
One of the biggest blockers to experimentation and innovation on Sol is distribution (or lack of).
Put simply, if your idea doesn't fit inside the box given to you by the big infrastructure (routing, indexing, etc), then no one will find your token/idea/launchpad etc.
Some have tried to counter this by launching a visible token (i.e via pump or wherever) and then directing that attention to their own site but that's still too high friction.
Both the expense of building, and the skill required to do so have been reduced 90%, and we should be using this to innovate and experiment. Not just rely on a few big corps to decide what features are 'worth it'.
So i'm building Pipeline (it will never have a token, and any potential revenue if it succeeds will be part of the @fffactoryhq ecosystem).
Pipeline is a permissionless onchain router for new bonding curves, AMMs, launchpads etc. Creators can register their contracts with Pipeline, and infra (trading bots, terminals, charting) that has Pipeline integrated will pick them up automatically.
Anyone that's interested in experimenting with the platform, looking to build something themselves, whether that's things to register with Pipeline, or infra built on top of it, please drop me a dm or comment here.
Currently only live on devnet as still running tests etc, will likely build a few fun AMMs/T22 hooks that can be experimented with myself to go with the mainnet launch, but anything extra built by others would be awesome (if you build your own amms, bonding curves, launchpads etc you can charge your own fees on them).
github.com/MetalLegBob/pipel…
Viktor retweeted
I've been talking about the @fffactoryhq for 7+ months now and yet everyone will end up saying
'ah I wish I listened to Cryo when it was still low'
Get away from the slop and do some research PLEASE
Perpetual financial motion
some ideas just off the top:
- t22 hook that restricts trading to a single pool until X amount of volume is released (prevents vampire 3rd part lpers in early stages) and allows the pool to grow. After volume goal is reached the hook gets nulled and token becomes composable everywhere
- an AMM that understands which token is which in an LP and takes the tax in the specific one everytime (then pair with a t22 that restricts trading to avoid being vamped)
- someone build a screener/trading bot/terminal (pls I hate doing front end stuff)
- some variation of Pool4 but for sol (if you do this please route at least some of the fees to adam/imd)
- pool that lends the unused parts of its LP out via Kamino or wherever
- 'single use' amms designed specifically for a new defi protocol/game tokenomics or whatever else you can think of
make onchain great again
Viktor retweeted
One of the biggest blockers to experimentation and innovation on Sol is distribution (or lack of).
Put simply, if your idea doesn't fit inside the box given to you by the big infrastructure (routing, indexing, etc), then no one will find your token/idea/launchpad etc.
Some have tried to counter this by launching a visible token (i.e via pump or wherever) and then directing that attention to their own site but that's still too high friction.
Both the expense of building, and the skill required to do so have been reduced 90%, and we should be using this to innovate and experiment. Not just rely on a few big corps to decide what features are 'worth it'.
So i'm building Pipeline (it will never have a token, and any potential revenue if it succeeds will be part of the @fffactoryhq ecosystem).
Pipeline is a permissionless onchain router for new bonding curves, AMMs, launchpads etc. Creators can register their contracts with Pipeline, and infra (trading bots, terminals, charting) that has Pipeline integrated will pick them up automatically.
Anyone that's interested in experimenting with the platform, looking to build something themselves, whether that's things to register with Pipeline, or infra built on top of it, please drop me a dm or comment here.
Currently only live on devnet as still running tests etc, will likely build a few fun AMMs/T22 hooks that can be experimented with myself to go with the mainnet launch, but anything extra built by others would be awesome (if you build your own amms, bonding curves, launchpads etc you can charge your own fees on them).
github.com/MetalLegBob/pipel…
Viktor retweeted
Hopefully at least 1 person tailed my conviction
Waiting for finn to endorse officially
The leaks showed bagspay having a "buy bagspay" button under their "app store" instead of "launch" like the others
+ he disabled every PVP possible with solana:H8WTvo3ZXSYHJAcYqzCforbG8coR1FDAX2kHFfqHBAGS
They tried to PVP it by sending bagspay X account Xmoney and it's the only coin where that Xmoney didn't want to go through lol
Also the buybacks changed from 20% bought back 80% held to 20% held 80% bought back
+
The buyback wallet manually aped 100 sol
I took some P, but I think this is lined for 10-20M+
Why I think @BagsPay is the only project that can succeed with X Money payouts
I've spent the last few days dding how memecoin fees -> X Money payouts actually work. Here's why others failed and why Bags will actually do it.
The first model (usePaid):
Tokens routed fees to a treasury -> the treasury claimed them ->paid in dollars to X accounts via X Money
Why it broke:
- Every payout came from one pre-funded X Money balance. Each dollar had to go crypto - exchange - bank - X Money. On a $1.5M day, the banking side couldn't keep up, then they paused the payouts
=CENTRALIZATION
- Converting crypto and sending dollars to third parties is money transmission. That requires licenses or a licensed partner. A business account from X wouldn't solve that alone.
=AML & KYC ISSUES
Worth noting: limits were never the issue. X Money's FAQ says P2P transfers have no limits once you're verified. The problem was banking, not X TOS.
How Bags does it ( as seen on their test token they launched few hours ago)
Fees - USDC on Base - P2P - get fiat - send to xmoney
Either finn would be the P2P on each side ( meaning he will buy every ask in fiat that he puts in USDC, or he leaves it fully P2P - organic)
And the biggest thing is, Finn can absolutely get the KYC & AML sorted, all you need is an outside market maker you pay a retainer, to run your offramps.
To close this off, it's just my findings/thoughts, why I think only Finn can succeed with Xmoney payouts and all other coins cannot succeed unless they have as big reputation with banks ( 10 fig vol on bags) + get MMs with licenses
I hope @finnbags succeeds with this, could go a very long way and achieve what paid tried to, but sustainable
The obvious way we could have built this:
Claim everyone’s fees into one treasury, sell them on an exchange, wire the dollars to a bank, top up one X Money balance, and pay creators from there with daily limits.
It works until it doesn't.
One balance. One bank. One point of failure.
Have we forgotten why we're here?
Tomorrow on MCG
$MOTION | @fffactoryhq w/@metallegbob
📅6th November, Friday
🕛12:30 PM EST
📍Tune in here
Presented by @fomo and @MetaDAOProject
Every coin launched on MEME OS now ends in "memeos."
pumpfun, Meteora, Raydium: same signature on every contract address.
One step closer to having our own identity onchain. When you see ...memeos, you know where it was born.
Want your own address? Custom CA is still there.
memeos.world

































