- Joined
- Jul 25, 2006
- Messages
- 12,147 (1.87/day)
- Location
- Nebraska, USA
System Name | Brightworks Systems BWS-6 E-IV |
---|---|
Processor | Intel Core i5-6600 @ 3.9GHz |
Motherboard | Gigabyte GA-Z170-HD3 Rev 1.0 |
Cooling | Quality case, 2 x Fractal Design 140mm fans, stock CPU HSF |
Memory | 32GB (4 x 8GB) DDR4 3000 Corsair Vengeance |
Video Card(s) | EVGA GEForce GTX 1050Ti 4Gb GDDR5 |
Storage | Samsung 850 Pro 256GB SSD, Samsung 860 Evo 500GB SSD |
Display(s) | Samsung S24E650BW LED x 2 |
Case | Fractal Design Define R4 |
Power Supply | EVGA Supernova 550W G2 Gold |
Mouse | Logitech M190 |
Keyboard | Microsoft Wireless Comfort 5050 |
Software | W10 Pro 64-bit |
I did read it. And it is still wrong. They may be unavoidable, but they surely are not neutral. If they were neutral, they would have zero impact on throughput. And zero impact on throughput is not a bottleneck.Bottlenecks aren't bad they are unavoidable, they are neutral. I already wrote that - again read my posts before responding to them.
That's right!A bottleneck is when a part of the pc isn't fast enough to hand over data to the other so it gets slowed down (eg slow cpu).
Of course that's a bottleneck. Again another tunnel vision perspective. If you don't have enough RAM and that forces the system to wait while the data is stuffed into the PF or read from the drive, for example, that's still a bottleneck. The speed of the RAM is just one factor but so is the amount. Bottlenecks in one area affect the whole throughput.A ram bottleneck isn't even that. A ram bottleneck is a missing quantity to load all data into it.
For example, using the freeway again - if a crash shuts down lanes causing a 27 minute delay in you getting home, it does not matter if the crash happened on the on-ramp, off-ramp, or anywhere in between. The end result is the same - you are still 27 minutes late getting home. Does it matter to your kids if the crash (bottleneck) was at the on ramp instead of the off ramp? No. The "end result" (throughput) is exactly the same.
No I'm not. A bottleneck is a bottleneck. You are trying to narrowly define a bottleneck as a delay that can only occur at only one or two places. That is not true! Bottlenecks can occur anywhere and affect the entire throughput - regardless where in the process they actually occur. That is exactly why too little RAM can create a bottleneck, just as too slow a CPU can, just as a slow drive, or slow internet connection.You're overstating what a bottleneck is.
Tell us, how is it better (or worse) if a 3 second delay for results to appear on the monitor are due to a slow CPU, instead of a slow GPU? How is this 3 second delay better (or worse - or different) than that 3 second delay?
I say no bottleneck is good. It does not matter where it occurs, it is bad, and all are unwanted. One is not better than another. It may be unavoidable due to budget restraints, Laws of Physics, or the technical limits of the state-of-the-art.
Last edited: