The ISU-152 idled for roughly 10 seconds almost directly in front of the Hetzer, yet somehow remained invisible
Instance 2: foliage.
The Panzershrek was INSIDE the bush and otherwise had an unobstructed view.
Instance 3: Smoke. As soon as the smoke cloud dissipates the Hetzer spots its target.
You can clearly see that smoke didn't obstruct the target. Even when the ISU-152 was covered with heavy smoke after being knocked out and pelted with grenades, it was STILL VISIBLE to the Hetzer.
Instance 4: There are two trees that are "trunks only" between the T-34 and the Hetzer. The T-34 got lucky and spotted the Hetzer before being spotted in return.
What a LOAD OF CRAP.
The Hetzer was completely incapable of detecting the T34 from when it slowly lumbered into view, right up to the moment that it was knocked out by the T34s third shot. FORTY-THREE SECONDS LATER.
* take two companies if infantry. Call it 150 eyeballs a side, and 150 targets. That's 150^2 potential spotting checks per cycle. For round numbers, let's say there's an average of 10 cycles per minute, so in an hour, 600 cycles times 22500 = 13.5 million potential checks per hour. And that's just one company a side. Many, many of those checks are handled "in summary" by the LOS table, but it's all part of a system that, generally, works.
What a LOAD OF CRAP.
Out in the real world, it's a simple matter of coding an algorithm that extrapolates a basic vision radius and checks yes/no to whether an enemy is visible.