TC4350 and TC7650Modem speed comparison?

Toggle sidebar Toggle sidebar
N

northon1

Prominent
Mar 9, 2017
13
0
510
Much has been posted about the Puma based modems having performance loss.
In my area, and with my preferred ISP, the speed packages are up to 250Mbps.
The modems that seem to be favoured are:-
- Technicolor TC4350 DOCSIS 3.0, 32x8 (Puma chip).
- TP-Link TC7650 DOCSIS 3.0, 24x8 (Broadcom chip).

Has anyone compared these modems and would the extra channels for bonding, on the TC4350, give any speed head room or would the Puma troubles just become more apparent?
For now the home isn't gaming. Just downloading and streaming of movies between 2 (3 max) users and possible multi web pages open at the same times of day.
The other (minor) thought is the TC7650 is 30% more to purchase right now and Giga speed is around the corner for the ISP, but it is unknown if the will be any tier added between the 250Mb and 1Gb.
 
Solution
B
Most the puma issue has been patched. You can still detected with specially crafted programs but it does not seem to cause issues like it did.

It really doesn't matter how many channels your modem has it only matter how many the ISP uses. Many ISP run only 16 channels when they claim 250mbps. 16 channels could run at 686mbps if they really wanted to but they seem to run simpler encoding. Most times the ones that use 24 channels are offering 350mbps.
Sort by date Sort by votes
B

bill001g

Titan
Aug 9, 2012
28,277
2,741
125,640
Most the puma issue has been patched. You can still detected with specially crafted programs but it does not seem to cause issues like it did.

It really doesn't matter how many channels your modem has it only matter how many the ISP uses. Many ISP run only 16 channels when they claim 250mbps. 16 channels could run at 686mbps if they really wanted to but they seem to run simpler encoding. Most times the ones that use 24 channels are offering 350mbps.
 
Upvote 1 Downvote
Solution
N

northon1

Prominent
Mar 9, 2017
13
0
510
I read a brief post implying the X86 chip was being over taxed. Do you know if the problem is to do with the physical channeling in and out of the chip, or the does the software forces the chip to handle the data in a way that that isn't efficient?
 
Upvote 0 Downvote
B

bill001g

Titan
Aug 9, 2012
28,277
2,741
125,640
I had puma chips all along and was not impacted by it so I did not read in detail. It was related to cpu load, there was discussion that traffic was passing through processor rather than just the modem part. Not sure by the time I even heard about it my ISP had patched my modem so I could not even test it.
 
Upvote 1 Downvote
N

northon1

Prominent
Mar 9, 2017
13
0
510
Up and running about 14 hours with TC4350 (32x8) on 75Mbps service (just to test difference from old 25Mbps service with SB6120 (4x4) modem) before seeing how the higher speeds perform.
Ping: 28 ms, Jitter: 1 ms, Download Speed: 75.9 Mbps, Upload Speed: 10.1 Mbps, Completed: Fri Feb 23 2018 11:48:43 GMT-0500 (wired network).
11 ms, 75.32Mbps, 10.07Mbps 23 2018 12:20
Jitter averages 4 ms with that particular test page both are by Ookla
 
Upvote 0 Downvote
B

bill001g

Titan
Aug 9, 2012
28,277
2,741
125,640
You can't read a lot into small difference. The server and the number of users running tests on the server can affect the numbers a small amount. Be 100% sure it is choosing the same server.

Pretty much as long as you get fairly consistent results from day to day it is fine. Speedtest is fairly artificial number, you can see this by choosing different servers even in the same city.

You only worry about jitter and ms if the numbers are large.
 
Upvote 0 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom