Question CS2 Spectator List — Post-Update Logic and Offset Fixes

[byte_corvus]

The recent update for CS2 just dropped and, as expected, it nuked common spectator list implementations. If you're relying on the old observer services traversal, you're likely hitting a wall or reading garbage memory.

It looks like the standard way of grabbing m_hObserverPawn and iterating through the m_pObserverServices to find the m_hObserverTarget is returning invalid handles or failing to validate the local pawn correctly.

The Current Broken Logic
Most pastes and even decent internals are using this snippet or a variation of it to detect who's watching them in Premier or matchmaking:

if (config.bSpectators)
{
    uint32_t obsPawnHandle = Game::Read<uint32_t>(controller + Offsets::m_hObserverPawn);
    if (obsPawnHandle != 0 && obsPawnHandle != 0xFFFFFFFF)
    {
        uintptr_t obsPawn = Game::GetEntityByHandle(obsPawnHandle);
        if (obsPawn && obsPawn != localPawn)
        {
            uintptr_t obsServices = Game::Read<uintptr_t>(obsPawn + Offsets::m_pObserverServices);
            if (obsServices)
            {
                uint32_t targetHandle = Game::Read<uint32_t>(obsServices + Offsets::m_hObserverTarget);
                if (targetHandle > 0 && targetHandle == localHandle)
                {
                    // Name extraction logic using m_sSanitizedPlayerName
                }
            }
        }
    }
}

What to Check
If your spec list is empty or crashing your overlay, start by dumping the latest schemas. Source 2 is notorious for shifting member offsets even in minor patches.

1. Verify m_hObserverTarget — Check if it still points to the correct entity index or if the handle encoding has changed.
2. Check m_pObserverServices — Sometimes the pointer chain breaks if the entity isn't fully updated in the cache.
3. Sanitized Player Name — If you're getting empty strings, the pointer at m_sSanitizedPlayerName might be moved.

Spoiler: Technical Note

Keep in mind that while some external solutions struggle with these shifts, real-time debugging with ReClass.NET usually reveals the new structure within minutes. Don't just wait for an offset dumper to update; look at the memory yourself.

While some are still trying to figure out why their public pastes are failing, Infocheats users should be verifying their entity iterator. If the pawn handle is valid but the target check fails, Valve might have tweaked how observers are registered internally to mitigate spec-list detection.

anyone found the new offset for m_hObserverTarget yet?