16 hours ago, Virtual-Chris said:

Here's an interesting benchmark for comparison...  This guy has a 5950X with a 3090 and he's getting over 60fps in photogrammetry heavy Tokyo.  With my CPU and the same GPU, I'm getting 40fps.  So throwing more CPU at this game will definitely help with photogrammetry.  Hmmm. 😧

https://www.youtube.com/watch?v=xxcTRDFivLA

 

Did the same flight, same settings,  with my setup, sans the obligatory flying between buildings, and same results.  62 FPS at most 52 at least, with the average I would guess around 58 FPS.  The plane was fun, and Tokyo is stunning, although I’m not fond of tail draggers.  

Good to know that amd is offering competition.

It brought another question to mind though. My monitor is 60 htz, yet it displays frame rates above 60. I thought I would be locked into 60 or less?

10 minutes ago, FrankR409 said:

Did the same flight, same settings,  with my setup, sans the obligatory flying between buildings, and same results.  62 FPS at most 52 at least, with the average I would guess around 58 FPS.  The plane was fun, and Tokyo is stunning, although I’m not fond of tail draggers.  

Good to know that amd is offering competition.

It brought another question to mind though. My monitor is 60 htz, yet it displays frame rates above 60. I thought I would be locked into 60 or less?

The game will generate more than 60 fps unless you cap it in some way.

Or are you referring to frequencies available in the windows display settings? It is actually possible to overclock a monitor. 

 https://www.windowscentral.com/how-overclock-your-pc-monitor

16 hours ago, Virtual-Chris said:

Here's an interesting benchmark for comparison...  This guy has a 5950X with a 3090 and he's getting over 60fps in photogrammetry heavy Tokyo.  With my CPU and the same GPU, I'm getting 40fps.  So throwing more CPU at this game will definitely help with photogrammetry.  Hmmm. 😧

https://www.youtube.com/watch?v=xxcTRDFivLA

 

just turn off ai traffic this helps alot, and i use low cockpit refresh rate too, i dont see the need for higher settings, you dont gain visualy anything, but more fps.

I am not going to do a full system build right now, but thinking of getting a 3070 for the next build. It will be bottlenecked for sure but still think i will pick up significant fps. When my FPS > 30 it is a great experience. Right now i fluctuate between 25 and 33. So i just need to get over the 30 fps hump. I bet a 1080Ti would do it but not future proofed and i was VR down the road. Decisions....

Edited November 14, 20205 yr by RobJC

23 minutes ago, RobJC said:

I am not going to do a full system build right now, but thinking of getting a 3070 for the next build. It will be bottlenecked for sure but still think i will pick up significant fps. When my FPS > 30 it is a great experience. Right now i fluctuate between 25 and 33. So i just need to get over the 30 fps hump. I bet a 1080Ti would do it but not future proofed and i was VR down the road. Decisions....

There is nothing particularly magic about 30 fps other than most people have 60hz monitors, so 30fps at exactly half of 60 works well as a vsync setting. 

My monitor is 144hz so I cap my game at 36 fps (exactly a quarter).

Whether a 3070 will be bottlenecked is going to depend mainly on screen resolution.  A 3070 by all accounts, would be expected to give very good frame rates at 1080p, be OK at 1440p but might struggle at 4K on ultra in dense areas and have no hope at 8K .

Edited November 14, 20205 yr by Glenn Fitzpatrick

Very good, a comparison with same CPU+ 6900XT would be interesting post Dec 8th

And in 1440p - where I suspect a full AMD setup will beat the RTX 3090 

Edited November 14, 20205 yr by hanhamreds

7 minutes ago, Glenn Fitzpatrick said:

There is nothing particularly magic about 30 fps other than most people have 60hz monitors, so 30fps at exactly half of 60 works well as a vsync setting. 

My monitor is 144hz so I cap my game at 36 fps (exactly a quarter).

Whether a 3070 will be bottlenecked is going to depend mainly on screen resolution.  A 3070 by all accounts, would be expected to give very good frame rates at 1080p, be OK at 1440p but might struggle at 4K on ultra in dense areas and have no hope at 8K .

Hey Glenn, not sure but on my system anything over 30 FPS is the key level. I have a 1440p G-Sync monitor. 

1 hour ago, FrankR409 said:

Did the same flight, same settings,  with my setup, sans the obligatory flying between buildings, and same results.  62 FPS at most 52 at least, with the average I would guess around 58 FPS.  The plane was fun, and Tokyo is stunning, although I’m not fond of tail draggers.  

Good to know that amd is offering competition.

It brought another question to mind though. My monitor is 60 htz, yet it displays frame rates above 60. I thought I would be locked into 60 or less?

Try it with overcast weather. Then we’re talking 30fps 🙂 

I watched another video earlier today, someone testing the R5 5600X vs the R5 3600 in Flight Simulator. GPU was 3090.

1440p: Up from 47 fps to 61 fps.

1080: Up from 49 fps to 62 fps.

4K: No difference.

I think for 1440p, the perfectly balanced system would be R5 5600X + RTX 3070 or 3060 Ti when it arrives. 

 

 

9 hours ago, Ianrivaldosmith said:

Same here, around 60 there too. Like I say it seems hyper threading make a big difference. 

Crazy that hyper threading would make that much difference.  When I bought my 9700k a year ago, the thinking at the time was that hyper threading and tons of cores really didn’t help with gaming, hence the existence of this part with 8C/8T. Now MSFS has completely changed that. I guess a new CPU is my next upgrade.  
 

edit: so looking into my options, it seems a 9900K (8C/16t) is a drop in replacement for about $400... but is only an upgrade in HT. Otherwise, I’m looking at a new motherboard and latest CPU for $1000 which would bring more cache, IPCs, etc. I wonder what is the best bang for the buck here. 

Edited November 14, 20205 yr by Virtual-Chris

Iff i could quickly chime in here, this thread leads me to see that gpu usage is fine even around 100, cpu usage should be substantially lower than that and hyper threading enabled for MSFS.(iff applicable).

Please correct iff that is not quite right, thx.

 

 

Edited November 14, 20205 yr by icewater5

11 hours ago, Ianrivaldosmith said:

Out of curiosity (as I have no clue), how quick do those instruments refresh in real life? 

Compare the rolling speedometer when you take off in both medium and High settings and you understand what I mean by jerky instruments.

Dubbel post

Edited November 14, 20205 yr by Ixoye

On 11/12/2020 at 4:23 AM, icewater5 said:

 

I thought hyperfocal had the same gpu at 4k res?.His system is running ok though. Unless he has another cpu

I’m running a 10900K at 5.2 GHz with a 2080 Ti. The 2080 Ti has the standard FE cooler with a very modest overclock (it becomes power limited very quickly). Using the Ultra preset at 4K I’m seeing about 45 FPS in the Mooney or Bonanza in most areas. Large cities such as New York have about the same frame rate as less populous areas (maybe 5 FPS less), and frame times are very consistent. The only issue I’ve had is for some reason Charles de Gaulle brings the sim to a crawl (~15 FPS). No other large airport does this.

The 3090 will not provide a massive performance improvement in MSFS, even in 4K until DirectX 12 is implemented.

6 minutes ago, Hyperfocal said:

I’m running a 10900K at 5.2 GHz with a 2080 Ti. The 2080 Ti has the standard FE cooler with a very modest overclock (it becomes power limited very quickly). Using the Ultra preset at 4K I’m seeing about 45 FPS in the Mooney or Bonanza in most areas. Large cities such as New York have about the same frame rate as less populous areas (maybe 5 FPS less), and frame times are very consistent. The only issue I’ve had is for some reason Charles de Gaulle brings the sim to a crawl (~15 FPS). No other large airport does this.

The 3090 will not provide a massive performance improvement in MSFS, even in 4K until DirectX 12 is implemented.

I went from a 2080ti to a 3090, and to me, it brought enough performance increase to warrant it. However, to each their own.