Comparison and justification of 'App-Script mode' versus 'Full Tunnel mode' with a VPS, focusing on performance, stability, and application compatibility.

If you get a VPS and it actually works you should just use that entirely and ignore mhrv-rs. Using this thing if you have a VPS is completely pointless and I have no idea why the upstream SOCKS feature or full tunnel mode have been added.

I think the logic behind Full Tunnel mode is that packets are first sent to Google’s servers and then forwarded to the VPS. This way, the connection cannot be interfered with at the local level, and the VPS IP won’t get blocked. Of course, the trade‑off is higher latency, but in my opinion, the increased server resilience makes it worth it. Correct me if I'm wrong.

> I think the logic behind Full Tunnel mode is that packets are first sent to Google’s servers and then forwarded to the VPS. This way, the connection cannot be interfered with at the local level, and the VPS IP won’t get blocked. Of course, the trade‑off is higher latency, but in my opinion, the increased server resilience makes it worth it. Correct me if I'm wrong. If that's the case then you're right, but I believe how the routing works is like this (I haven't been following this repo for a while so don't count on me): Anything that GAS can handle (HTTP(S)) is sent to it while everything else is set to the configured proxy (VPS in this case).

> I think the logic behind Full Tunnel mode is that packets are first sent to Google’s servers and then forwarded to the VPS. This way, the connection cannot be interfered with at the local level, and the VPS IP won’t get blocked. Of course, the trade‑off is higher latency, but in my opinion, the increased server resilience makes it worth it. Correct me if I'm wrong. > mode : Full client > app script > node > internet این درسته و من ران کردم فول تانل به همین صورته اپستریم ساکس ولی ی چیز کسشعره

Hi @ipvsami — fair question, here's the honest comparison so you can decide before spending on a VPS: **Latency (the biggest difference):** | | Apps Script mode | Full Tunnel mode | |---|---|---| | Per-request RTT | **2–7 seconds** | **300–800 ms** | | Reason | Apps Script execution time + Google's per-request overhead | Just Apps Script transport (the data hop is direct VPS↔origin) | This is *the* reason Telegram Web feels janky in apps_script mode. Every message exchange has at least one round-trip; at 2–7 s per round-trip, "is the message delivered yet?" turns into a real question. In Full mode, Apps Script is just the transport (the ~6 s execution-time cost still applies for connection setup), but **once a session is established, the data path skips Apps Script entirely** — TCP segments flow directly between your client → VPS → origin. Real-time apps suddenly feel real-time. **Throughput:** - **apps_script:** capped by the number of deployments × 30 concurrent fetches × 2–7 s ...