If front beam forming is employed, the room acoustic indeed really matters. Front beam forming to focus on front speech, as a technique employed by many hearing aids, is only effective if the room acoustic does not produce a lot of reverberation, like a room that is very large, and particularly with very tall ceiling as to bounce sounds around too much. The other conditions for front beam forming to work well is that the speaker you’re trying to listen to is not further than about 6 feet from you, and the noise has to be from your sides and your back and not also from the front. Of course you will hear speech in front well, but if the other folks who are talking but are not in front of you, you will not hear them well.
Did you find whether the Oticon Intent 1 any better in this respect? The idea is that the DNN should rely less on directional frontal beam forming to improve speech. It uses a different way to reduce signal to noise ratio contrast on the surrounding speeches.
However, one challenge that you have is that your hearing loss is in the severe to profound range, and is quite flat across most frequencies. This make the wide dynamic range compression very, very tight, with little room to maneuver. Below is an example that explains what wide dynamic range compression looks like. But imagine that your impaired range of hearing (the vertical red plane on the right side) is extremely narrow, even much narrower than what the example here is shown. Yours is actually in between about 80 or 90 dB to 120 dB, or only 30 to 40 dB of range. In contrast, the normal hearing range as seen in the left red square has a full dynamic range of 120 dB.
So you have only about a quarter of the dynamic range of what a normal hearing person has. This does not leave much room to discern acoustical differences between the various sounds. In your case, all the sounds get crammed into a quarter of the space (compared to a normal person), so they compete fiercely with each other for your attention. It makes it at least 4 times harder (if we’re being linear here, although it may not be linear and might be even worse) for your brain to be able to discern the differences between these sound characteristics and be able to tell them apart, because much of the nuances of the differences are lost by being squished in together way too much.
An analogy is that instead of being presented with a big canvas of a picture, now the canvas has shrunk to a quarter of its size. If there is only a few things on the canvas, you can still probably be able to make out those things OK if they’re shrunk to a quarter of a size. But if the canvas is a very busy picture with hundreds of items on there (analogy of a noisy places with hundreds of different competing sounds), if shrunk to a quarter size, it’s going to be much harder to focus and discern enough to be able to identify all the hundreds of those items individually, because it becomes more like a bowl of spaghetti on a much smaller canvas.
So in some respect, the front beam forming to focus on speech in front only might still be the best way to help with your type of hearing loss, because at least there’s only 1 speech in front painted on the canvas for you to see, while the other speeches are faded out. The Oticon open paradigm might just introduce just way too many pictures on that tiny canvas that no matter how clear they can make those tiny pictures out to be, it might still just be way too much details on a tiny little canvas for you to discern between them.
Hope this helps explain the challenges that your hearing loss faces, and maybe why something as good as the Oticon Intent may still not be able to surpass the approach that the Phonak hearing aids can deliver for you.