I can't claim specific numbers since like all benchmarks it depends on usage, but I am quite confident our speeds will impress based on how we handle sync. We d

I can't claim specific numbers since like all benchmarks it depends on usage, but I am quite confident our speeds will impress based on how we handle sync. We do not send the object state, but the specific operations instead. In addition, we send in binary format and compress with zlib, so all in all we are very optimized.